nglish
O p
anguage verseas
Ee,spec"ves and
nquiries Vol. 15, No. 2 (2018)
Journal Editors: SMILJANA KOMAR and MOJCA KREVEL
Of
University of Ljubljana
FACULTYofARTS
Ljubljana University Press, Faculty of Arts Znanstvena založba Filozofske fakultete Univerze v Ljubljani
Ljubljana, 2018
Contents
PART I: LANGUAGE
Nikola Dobric_9
Reliability, Validity, and Writing Assessment: A Timeline Zanesljivost, veljavnost in ocenjevanje pisanja: časovni pregled
Vesna Lazovic	25
How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and Serbian in Light of Relevance Theory
Kako se igrati/plačati z besedami? Raba besednih iger v spletnih bančnih oglasih v angleščini in srbščini v luči teorije relevantnosti
Jakob Lenardič	45
The Middle Construction in Minimalist Syntax: English and Slovenian Srednjiška zgradba v minimalistični skladnji: angleščina in slovenščina
Frančiška Lipovšek	63
The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene Fictive Motion Expressions
Vloga glagolov premikanja pri opisu poti v angleških in slovenskih izrazih fiktivnega gibanja
PART II: LITERATURE
Begona Lasa Alvarez	81
Mary Hays, an Eighteenth-Century Woman Lexicographer at the Service of "the Female World"
Mary Hays, leksikografinja iz osemnajstega stoletja v službi »ženskega sveta«
3
E L O P
PART III: ENGLISH LANGUAGE AND LITERATURE TEACHING
Nataša Puhner, Mateja Dagarin Fojkar	97
Developing EFL Writing with Year 5 Pupils by Writing Summaries Razvijanje pisnih spretnosti v tujem jeziku pri petošolcih s pisanjem obnov
PART IV: LIST OF CONTRIBUTORS
5
m • • ••
•7*< •.'ft
■K' • • •

E L O P
Nikola Dobric	2018' VoL .15j2), 9-24/!20)
TT . . rT„ r . .	revije.tt.uni-lj.si/elope
University of Klagmfott, Austm	doi: io.4312/elope.15.2.9-24
UDC: 37.091.26:808.1(091)
Reliability, Validity, and Writing Assessment:
A Timeline
ABSTRACT
Looking at the issue of validity and test validation, the historical and the theoretical progression has been well described both when it comes to educational assessment in general and language assessment in particular. A clear progression can be seen starting in the 1920s and culminating in the late 1980s/early 1990s (with minor notable developments since), and it is an advancement motivated and driven almost solely by new theoretical and practical considerations. Securing validity and validation with regard to writing assessment in particular, however, took a more winding route and was primarily shaped by a power struggle between externally administered standardized testing (and the supporting administrative bodies) on one side, and the practicing teachers of writing at higher education institutions on the other. The paper at hand outlines this evolution and gives a timeline of the events and major developments that have fueled it and explores the cutting edge of today.
Keywords: language testing; writing assessment; historical development; overview; validation; validity; standardization
Zanesljivost, veljavnost in ocenjevanje pisanja:
časovni pregled
POVZETEK
Vprašanje veljavnosti in vrednotenja testov je bilo z vidika zgodovinskega in teoretičnega razvoja že dobro opisano tako na področju splošnega pedagoškega ocenjevanja kot tudi na specifičnem področju ocenjevanja jezikov. Začetek razvoja sega v dvajseta leta preteklega stoletja in doseže vrhunec v poznih osemdesetih oziroma v začetku devetdesetih let 20. stoletja (z manjšim opaznim napredkom tudi po tem času). Gibalo razvoja so predvsem nova teoretična in praktična dognanja. Zagotavljanje veljavnosti in validacije ocenjevanja pisanja pa je potekalo po bolj ovinkasti poti in predvsem pod vplivom boja za premoč med zunanjimi standardiziranimi testi (ob podpori administrativnih teles) na eni strani ter visokošolskimi učitelji pisanja na drugi. Prispevek prikazuje ta razvoj in podaja časovni pregled dogodkov in večjih dosežkov, ki so ga omogočili, hkrati pa tudi raziskuje današnje najsodobnejše smernice.
Ključne besede: jezikovno testiranje; ocenjevanje pisanja; zgodovinski razvoj; pregled; veljavnost; validacija; standardizacija
MME1	LANGUAGE 9
Reliability, Validity, and Writing Assessment: A Timeline
1 Introduction - Indirect Assessment of Writing
If we leave aside the writing examinations one reads of starting in Ancient China almost two millennia ago (as a part of the rigorous assessment of civil servants underway during the Han Dynasty in the 2nd century AD) as too far removed contextually and in time, and somewhat historically inaccessible, we can say that the earliest more contemporary records dealing with the assessment of writing in the Western countries emerged in the mid-19th century. At the time, the turn was being made from the Western tradition of open debate (stemming originally from the Ancient Greek philosophical tradition) towards a more standardized model of uniform examinations (actually conceptually influenced by the previously mentioned Chinese tradition). In essence, across the Western world universities usually favored giving oral-based examinations, a tradition stemming from the Socratic approach to higher education. However, because of the displacement created by the Industrial Revolution, school-age children were suddenly being forcibly put behind school desks (a result of new compulsory education laws), a development which demanded a faster and more resource-friendly way to test larger numbers of students. One of the first actual records of written examinations coming up within a national dialogue is one by Horace Mann calling for written tests to replace oral examinations in Boston schools in 1840 (Huot, O'Neill, and Moore 2010, 496).1 In the 1840s oral tests were the standard in American schools, and the move was to introduce written tests in order to facilitate objectivity and reduce bias. Additionally, he also wished to build a transparent (standardized) system according to which one could compare the quality of different schools and teachers. As a result of this and similar calls, written tests were relatively quickly introduced across the United States, such as for example the Harvard's Exams in Writing in 1874. With this a new writing component was added to the Harvard University entrance exam, which included a short, speeded essay rated by the resident teachers, and was based on 'such works of standard authors as shall be announced from time to time' (Hobbs and Berlin 2001, 252). However, not long after the launch of the first essay-based writing exams at American universities came critical voices. On the one hand, the criticism focused on unreliability and subjectivity in terms of rating the essays (ironically enough, the same criticism that motivated the switch from oral examinations in the first place), and their lack of practicability and usability in a situation of increased numbers of students to process, on the other hand. Already in 1880 we have first publications highlighting the problems of unreliability that were inherent in the early essay-based writing examinations (Huddleston 1952). For example, the study published by Hopkins in 1921 demonstrated that the score a student received as a part of the College Board exam statistically depended more on the rater and on the administration conditions (when and where the exam took place) than on the actual writing performance, a situation also found in many European countries administering writing tasks as part of high-stakes examinations (Nikolov 2009). The Certificate of Proficiency in English (CPE) was instituted in London in 1913 by Cambridge Assessment (an organization that went on to become the leading international English language examining institution in the
The history of the development of standardized testing and its influence on writing assessment is best documented within the American framework, and hence the brief historical overview will mostly follow the story in this context, with a note that the development in other Western countries took a similar form with the exception of the UK, where the emphasis on direct methods of writing assessment and a focus on (context) validity was always more present.
10 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E L O P
world), and with its emphasis on essay (composition) writing was to face similar criticism. These very real issues were further emphasized not just by a continuous academic scrutiny (in a number of studies, such as Starch and Elliott (1912), Hopkins (1921), Sheppard (1929), and others), but also by both demographic and political developments impacting higher education in the USA at the time. Demographically, the numbers of pupils and students in the US rose exponentially, because starting from 1852 the different states began adopting the recently approved universal (and obligatory) public education laws. Politically, this fueled the foundation of the college Entrance Examination Board (cEEB) in 1899 as a not-for-profit organization aiming to expand access to higher education. And while this did mark a very important development, certain actions regarding organization of assessment within higher education (such as the centralization and outsourcing of language testing) created a massive power struggle and by extent had a great impact in terms of how validation of writing assessment developed in the subsequent century.
Backed by academic publications pointing out the severe problems of the unreliability of essay-based assessment of writing, weighed on by the increased numbers of students in need of processing, under pressure from high school representatives for a more standardized arrangement of entrance examinations into universities, and somewhat swayed by internal power issues, the decision was made by the cEEB to replace the then-perceived unreliable direct writing assessment (essays) with a much more reliable (though, of course, much less valid) indirect method. Unlike the US, the socio-economic context in the UK did not demand such an increase in processing power, and was able to maintain a much larger focus on context validity and maintain (for cPE unwaveringly for more than a century now) direct assessment of rating as a predominant method (Weir, vidakovic, and Galaczi 2013). Nevertheless, assessment of writing in higher education institutions in the USA (and many other countries) became as of that point almost solely conducted in the form of multiple-choice tests and, to a large extent, mostly outsourced to private institutions as well. This was to remain so up until the 1960s, as the issue of unreliability came up periodically and repeatedly in academic publications as the major factor in favor of indirect methods (such as Traxler and Anderson (1935), Stalnaker (1936), coward (1952), and others). Fortunately, this emphasis on continuously pointing out unreliability of direct writing assessment was to backfire in 1961 with a notable publication by Diederich, Frech, and carlton. Their study dealt with the problems of readers (raters) not being able to agree in terms of rating writing performances, identifying that as the major source of error when it comes to score stability (reliability) and, ultimately, fairness of direct writing examination. In brief, they asked 53 raters to score 300 writing performances on a 9-point scale, with the results of 94% of papers receiving at least seven different scores (and the inter-rater agreement peaking at only 0.31). Looking from today's perspective, we can easily see the flaws in the study as readers were bound to disagree seeing that they worked without any guidance, common training, or shared point of reference (as was pointed out even at the time in Braddock, Lloyd-Jones and Schoer (1963)). However, the results of the study regarding the inter-rater (dis)agreement are not the reason why this earlier work is to be considered so relevant. This importance actually stems from the fact that it got the academic community thinking about what criteria raters actually focus on when rating written work. This is because the study, apart from looking at the agreement, also collected some eleven thousand introspective reports from the raters, revolving around the features they were looking at when rating the performances (coming up, via factor analysis, with ideas, form, flavor, mechanics, and wording as the most salient features). This particular aspect of looking beyond just inter-rater agreement as an expression of reliability into issues such as the features of writing performance that influence rating, rater scores, and rater training and experience, was to establish a new base-line for researching writing assessment that was to mark
MME1	LANGUAGE 11
the next stage of the development in the field (Huot, O'Neill, and Moore 2010, 502). Studies moving towards establishing proof that direct writing assessment can produce reasonably reliable scores (such as Godshalk, Swineford, and Coffman (1966) and their outline of holistic scoring, for example) further opened the door to a generally increased focus on direct writing assessment (but not, as we shall see, to a much needed shift of focus from reliability to validity).
2 Direct Assessment of Writing - Focus on Reliability
Despite these developments and motions, the CEEB fought back all the way up to the late 1980s to preserve multiple-choice tests as the foundation of standardized (and usually externally managed) assessment of writing at universities and colleges in the USA (such as the well-known TOEFL examinations). The struggle was, however, futile, as the push in the other direction (towards direct examination of writing proficiency) was once more coming both from political and academic quarters (Breland 1983, 2). Politically, one of the issues was that the overwhelming focus on indirect assessment of writing was extremely worrying, because it technically led to hardly any writing being taught within the American educational system by the mid-1970s (Applebee 1981). Another political issue was the pressure coming from the English Departments at higher education institutions across the country. Rankled by both the feeling of disempowerment caused by the prescribed external testing services and the feeling that multiple-choice testing of writing did not really conform to their experiences as teachers of English and writing of what writing ability represented, a firm stand was made to change things. Within the US, we can trace the first major battle being fought in 1971 and the English Departments at California State University, the staff successfully rejecting the institutionalization of an externally administered multiple-choice test for their first year English equivalency exam (White 2001, 308), following this up with a creation of their own locally administered examination. Academically, with authors such as Paul Diedrich changed their stance towards direct writing assessment, and similar voices coming collectively from other publications (such as Brown (1978), Coffman (1971), Cooper (1977), Cooper and Odell (1977), and others), the move was a clear one towards the abandonment of the indirect method of assessment of writing ability. Having a wealth of information on where the source of reliability problems is to be found when it came to rating written performance, the two main focuses being on rater inconsistency and sampling bias, the majority of studies in this period revolved around the attempts to address these. With a better understanding of the rating process and what it entailed, there was a rise in research on scoring (such as that related relating to analytic scoring with Diedrich (1974), holistic scoring with Godshalk, Swineford, and Coffman (1966) or Cooper (1977); primary-trait scoring with Mullis (1980); or syntactic scoring with Hunt (1977); writing scales in Jacobs et al. (1981); Huot (1990) or Ericsson and Simon (1993); rater training and agreement with McIntyre (1993) or Weigle (1994); and other facets of score stability).
The fact that the socio-economical and academic conditions were favorable for direct assessment of writing to be promoted meant that by the mid-1980s most universities (and their language departments) were already implementing locally administered direct tests of writing ability in much the same way. This was a very positive step, as it allowed for tests to be developed by the faculty that worked together within the same program for their own purposes, with their own goals and within the shared system and shared curriculum, and yet also able to maintain reliability at an acceptable level. However, as we mentioned, while the switch to performance-based testing of writing was a very important development, what did not change (but should have) was the focus, which needed to shift from practicability of testing and reliability issues to much more important questions of aspects of validity unrelated to scoring (mostly present within the UK tradition, though).
12 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E L O P
3 Direct Assessment of Writing - Focus on Validity
Validity in terms of assessment indicates that the results obtained from the given measurement procedure objectively reflect the phenomenon the said procedure is intended to measure, and that the measurement at hand has not been obtained due to any measurement-irrelevant factors or chance). In this respect validity depends on the degree to which quantified measurements of presumptive behavior or ability are blurred by other factors (Sigott 2004, 44). Different from reliability, which can be seen as the quality of the data collected, validity is the quality of the inferences (and decisions) we can make following the measurement (Chan 2014, 9). In this sense validity also depends on the degree as to which aspects of the said behavior or ability which the test is supposed to measure are covered by the given test (Sigott 2004, 44). Validation is the process in which we gather and evaluate evidence to support the said appropriateness, meaningfulness, and usefulness of the inferences and decisions we make based on measurement scores (Zumbo 2007; 2009). Unfortunately, one of the products of the described power struggle between the (external) testing providers (who were seen as one of the actors advocating the indirect writing examinations) and the public bodies (such as the CEEB) on one side and the researchers and practicing teachers of writing (who felt unheard and disempowered) on the other, was that writing assessment did not catch up with the theoretical developments revolving around validity that had actually been under way in educational and language testing since the 1950s. Although there were voices calling for more insight into the content of the tests (the likes of Wiseman (1949) or Wiseman (1956)), most of the research on the assessment of writing taking place between 1960 and 1990 was into the reliability of the direct methodology of writing assessment (the same as during the previous stage of development regarding indirect methodology). The focus did not appropriately really shift towards validity until the early 1990s.
The earliest accounts of validity considered within the framework of language testing can be traced to authors such as Lado (1961) and Davies (1968)2. Their account of validity focused on the face values of the test (the so-called validity by assumption), on the content of the test, on control of extraneous factors, on conditions required to answer test items, and on empirical insight (D'Este 2012, 63). At the same time, Campbell and Fiske (1959) described the validity of language tests as having a convergent dimension (measures that should be related are perceived as related) and a discriminate dimension (measures that should not be related are found to be not related). Campbell and Stanley (1966), as one more early account, featured the concepts of internal and external validity. It is only decades later that we can see perhaps the most influential account of validity within language testing, that of Bachman (1990). It was written as a reaction to the unified theory of validation famously put forward by Messick in 1989, which Bachmann embraced, but went on to identify three factors that support this overall validity (D'Este 2012, 67): content relevance and content coverage (what is known as content validity); criterion relatedness (i.e. criterion validity); and meaningfulness of construct (construct validity). Content validity here refers to the domain specifications which underlie the test, criterion relatedness refers to a meaningful relationship between test scores and other indicative criteria, while construct validity relates to the extent to which performance of the test is consistent with the predictions we make on the basis of the theory of abilities (Bachman 1990, 246—69). In addition, Bachman also imported the consequential (or ethical) basis of validity from Messick and other authors, which
2 While the earliest considerations of validity within psychological and educational settings trace back earlier, with beginnings in the 1920s (and authors such as Scott (1917), Thorndike (1918), Ruch (1929), or Tyler (1934)), with the major movement here starting in of the 1950s (and the work related to the first Standards of Educational and Psychological Testing of 1955).
MME1	LANGUAGE 13
refers to the fact that tests have not been designed to be used in an academic vacuum but rather have real-life applications and are influenced by society as a whole. The final important notion, as argued by Bachman, was of the inclusion of the concept of reliability within validity itself. His argument was that when it comes to language testing of any kind (and it so especially in the case of writing assessment), it is not easy to distinguish between effects of different test methods or between traits and test methods (1990, 239).
Stemming from Bachman's account, the 1990s and early 2000s saw several related discussions and models of validity in language assessment, such as Kane (1992; 2001), Sigott (1994), Alderson, Chapman and Wall (1995), Miller and Linn (2000), and Wier (2005). If we were to take as an example, as the most recent and most language-assessment-related account, Weir's socio-cognitive model of evidence-based test validation (2005, 17—37), this presents overall test validity as an interplay between five different types (discussed in more detail further on): influenced by test-taker characteristics, there are cognitive validity, context validity, scoring validity, consequential validity, and criterion validity. cognitive validity has as its emphasis the determination of the cognitive processes which are to be used as a model for designing test items. Context validity follows the idea of moving away from the sole focus on linguistic representation and including the social and cultural dimension within which the writing performance has been produced (Weir 2005). Scoring validity, in line with Bachman's already discussed elaborations on reliability belonging within validity, focuses on all aspects of the assessment that could have an impact on the scores. Finally, criterion-related validity in this account keeps the focus as traditionally established, and revolves around a comparison to any reliable external measurements.
4 Direct Assessment of Writing — Discussion
What we can see if we dissect all of the accounts presented above, from Bachmann onwards, is the contemporary view of validity and validation in language testing understands that validity is a unified concept (which includes reliability), and is an aspect of a test's use to be regarded as a whole. However, we can also see a practical need for a focus on particular individual aspects (types) of validity that should be covered. These individual aspects reflect sources of evidence for validity that we can tap and steps to be taken within any validation effort (regardless of the theoretical preference of the terminology marking the given different aspects. This is clearly evident in terminological disagreements such as construct validity vs. validity, division on content, criterion, and consequential validity, and so on). Applying this kind of a practice-driven (authenticity-driven) model to direct writing assessment, we can start with understanding it as being strong or weak (McNamara 1996). Direct writing assessment in a strong sense incorporates tasks which represent real-world performance and are judged according to real-world criteria, the focus being on the successful fulfilment of the task and not (only) on language proficiency. Direct writing assessment in a weak sense puts the focus on language use, where the task does indeed resemble a real-world purpose, but the goal is only to extract a writing performance for the purpose of ascertaining language competence. It is a distinction that influences everything from the construct to the scoring, and it is safe to say that in most cases writing assessment in higher educational setting takes the form of the latter (Tsai 2002, 1). Once a choice is made in this direction, further conceptualization (and retroactively understanding) of the construct (including all other elements of validity as well) comes essentially at one of the three stages (Bachman and Palmer 1996; Weigle 2002): design stage, operationalization stage, and test administration.
14 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E L O P
The design stage is extremely important for ensuring validity and it should, according to McNamara (1996, 43), revolve around sampling of the task from the communicative situation the test is to be a proxy of — this would include consulting expert informants, examining available literature, analyzing and categorizing communicative tasks, collecting and examining relevant texts, and deciding on a broad test method (Tsai 2002, 2). In the operationalization stage information gathered in the design stage is transformed into concrete test specifications and detailed procedures the test takers and readers (raters) are to follow. These specifications should generally include the description of the test content, criteria for correctness, and a sample of tasks (Douglas 2000), and comprise of scoring rubrics, writing prompts, rating scales, and similar (Weigle 2002). Finally, the test administration phase focuses on the actual tests being administered, where the validation revolves around the data that stem from the elicited writing performances (such as the scores and the washback information), while the stage is firmly footed in different statistical methods. This division into practical stages of test development and administration is useful, as it points to the different steps which should be taken at different stages in order to ensure the validity of a particular interpretation and use of a particular test, and to the fact that a useful way to observe the process of validation is to imagine it as a priori (before the test administration) and a posteriori (after it has been administered). Additionally, it is useful to observe validity in terms of the types of evidence that support it, along the lines of Bachman's division into evidential and consequential bases of validity (1990, 248)3. Finally, the interplay between validity and reliability is also worth discussing, together with how Bachman (following other similar voices such as Loevinger (1957), Messick (1989) or Cronbach (1990)) argues that the two are in fact complementary aspects of identifying, estimating, and interpreting different sources of variance in test scores (1990, 239).
4.1 A Priori Validation
Unlike the subsequent stage of a posteriori validation that is largely grounded in quantitative data and statistical methodology (and which is usually feared and avoided by practicing teachers), a priori validation efforts are usually conducted on a regular basis by most practitioners. The reason is that a priori validation takes places during the test compilation phase, and most commonly includes the very logical, common-sense, steps anyone seriously compiling an assessment tool would think of (though it can also incorporate some statistics-based operations). A priori efforts basically focus on the commitment of the test designer to make sure that he or she understands the behavior or performance which is the target of the assessment, that the content of the test is relevant to the real-life behavior or performance at hand, and that the criteria of success (correctness) are clear both to the test-taker and the test-rater. This stage of validation is hence much less opaque to non-experts and, as indicated, covers the stages of test design and test operationalization.
Starting always with the understanding of the behavior or performance of interest, the first step in test design (and a priori validation) is the consideration of construct validity in the narrow sense. As we noted previously, a construct can be seen as a definition of people's attributes that are assumed to be reflected in their performance (Cronbach and Meehl 1955, 283). This means that to properly measure the 'extent' of someone's writing ability we need to understand the nature of that ability as it is, in our minds, and we need to be able to describe it in a sufficiently
3 Evidential basis of validity, according to Bachman, is grounded in evidence that supports the relationship between test scores and their interpretations and subsequent use (1990, 248). The consequential (or ethical) basis of validity refers to the fact that tests have not been designed for use in an academic vacuum, but rather have real-life applications (Bachman 1990, 280).
MME1	LANGUAGE 15
clear manner. Hence, construct validation is related to the basic question of what is the nature of that something that an individual possesses or displays that is the object of our measurement (Messick 1975, 957). The part that takes place a priori (sometimes referred to as the logical part of construct validation) involves first defining the construct (the ability) theoretically, or rather scouring the existing literature on descriptive accounts of writing ability. Accounts such as these try to capture the inner-workings of the cognitive processes at the basis of what we could call 'an ability to write' (and often perhaps the 'ability to write in a foreign/second language' as well). Sometimes this is quite challenging, as writing assessment as a discipline has not really excelled at writing about writing (Huot 1990), but rather at the more practical work on assessment methodology (such as writing scales and rater training procedures). That is why there is a feeling that theoretical accounts of writing competence as a phenomenon (unlike accounts of language competence in general) are somewhat lacking, and that more work still needs to be done in this direction (Yancey 1999). Nonetheless, finding (or setting up) a comprehensive and meaningful theoretical account of the construct in the design stage allows us to perform many actions which can assure subsequent high levels of validity. Understanding the construct well allows us to identify the sources for the most meaningful sampling of test content, and to identify which test method would be best to elicit behavior (performance) which would most resemble the construct at hand and the relevant real-life performance. These developments we would carry over to the operationalization stage, where the logical (theoretical) part of construct validation provides us with means of linking the actual ability we are measuring with the content of the test and with the scores that end up quantitatively representing our measurement. Adjacent to this understanding of the ability itself is the insight into the metacognitive strategies (or strategic competences, as termed in Bachman and Palmer (1996)) seen as mediating between the trait and context, and comprising of faucets such as goal setting, assessment of needs to achieve the purpose, planning, monitoring, and organization of language and topic — all contributing, in fact, to what Wier refers to cognitive validity (2005, 86). In terms of writing assessment, Shaw and Weir (2007, 34) recognize six different aspects of cognition behind writing ability:
•	macro-planning: gathering of ideas and identification of major constrains such as genre, readership, and goals;
•	organization: ordering the ideas and identifying relationships between them;
•	micro-planning: focusing on individual parts of the text and considering issues such as the goal of the paragraph in question, including both its alignment with the rest of the text and the ideas and the sentence and content structure within the paragraph itself;
•	translation: the content previously held in a propositional form is transferred into text;
•	monitoring: focusing on the surface linguistic representation of the text, on the content and the argumentation presented in it, and its alignment with the planned intentions and ideas; and
•	revising: results from the monitoring activity and involves fixing the issues found to be unsatisfactory.
Alongside the cognitive aspects behind writing as a performance, Weir also emphasized the need to consider the test-taker actively in terms of personal individualities. He distinguished between three classes of test-taker characteristics (Shaw and Weir 2007, 5):
•	physical/physiological characteristics, which include any special needs on the side of the test-taker, such as those stemming from dyslexia or eyesight impairment;
16 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E L O P
•	psychological characteristics, including test-taker motivation, personality type, learning styles, and more; and
•	experiential characteristics, incorporating factors such as the degree of test-taker familiarity with the test format or content.
These individual characteristics can then be viewed as systematic, if they affect a test-taker's performance consistently (such as dyslexia or personality traits) or unsystematic, when they have a random, perhaps one-off effect (for example, motivation or test format familiarity).
Once we have a reasonable grasp of the nature of writing ability (and a complete grasp in theoretical terms is pretty hard to achieve when it comes to social sciences and humanities), the test-takers' characteristics at play, and the cognitive processes taking place in our minds while writing, then while still in the design stage we need to tackle the context surrounding the test. Related to the social and situational background (Weir 2005, 57), this revolves around the task setting and conditions necessary to ensure appropriateness of the test content, clarity and conformity to the construct, the intended use, and the intended stakeholders (Shaw and Wier 2008, 63):
•	setting = task: includes the expected task format (consideration the genre), the purpose (expected form and communicative function or the nature of information in the text), the clear knowledge of the criteria (task instructions), the weighing (focusing on the relative contribution of the different parts/aspects of the test), the text length, the constraints (i.e. speededness, additional resources, and more), and the expected writer-reader relationship (addressee information);
•	setting = administration: incorporates physical conditions (venue, background noise, lighting, and more), uniformity of administration (same specifications for all test takers), and the security involved (controlled access); and
•	linguistic demands = text input and output: the focus is on the lexical resources (vocabulary), structural resources (morpho-syntax), the discourse mode (considerations of genre, rhetorical task, and pattern of exposition), the functional resources (referring to the fulfillment of the intended communicative purpose of the writing), and the content knowledge (background and the subject matter knowledge).
Also important to mention here is that, unlike most of the other a priori validation efforts, ascertaining validity of contextual (content) features such as the appropriateness of the lexical resources, structural resources, or the discourse mode can be undertaken empirically (quantitatively) as well.
The last a priori puzzle-piece in the jigsaw that is validation (and ultimately the degree of validity) of a particular issue of writing assessment is in the operationalization stage, where we define instructions on how to, essentially, obtain reliable scores. Called scoring validity by Weir (2005, 117), Shaw and Weir list the a priori considerations as follows (2007, 146):
•	criteria and type of the rating scale: focuses in essence on the marking scheme and the different approaches to assigning a number to the measurement (i.e. primary trait scoring vs. holistic scoring vs. analytic scoring);
•	rater characteristics: identified as one of the biggest causes of score variability when it comes to writing assessment; the rater is observed both in terms of the rater-candidate and the rater-item interactions (again dividable into physical/physiological, psychological, and experiential characteristics);
MME1	LANGUAGE 17
•	rating process: different prescribed rating procedures (such as for example the principled and pragmatic two-scan approaches, the read through approach, or the provisional mark approach);
•	rating conditions: includes the setting (on site or at home, for example), the medium of the writing performance (handwritten vs. electronic), time constraints (deadlines), and scaffolding (ways in which examiners are advised); and
•	rater training: perhaps the most crucial aspect in achieving higher levels of inter-rater agreement (lowering the influence of severity and other effects that stem from the rater).
After having completed all of listed actions, and ultimately having arrived at a version of the test that in its content and specification reflects entirely the theoretical description of the behavior (ability) we are interested in measuring, it is time to administer the test and then see how much the results (scores) obtained do in fact conform to the expectation of validity we have built up in the design and operationalization stages.
4.2 A Posteriori Validation
One of the first aspects of validity that most research into test interpretation employs after the test has been administered is generally termed criterion validity. As indicated, it revolves around correlations with other comparable measures of the same ability. This is one of the go-to procedures in smaller-scale validation efforts, as it is simpler to set up in comparison to other procedures — all one needs is an established comparable existing (or future) measurement which is already considered as valid, to which you can make comparisons in terms of obtained measurements (scores). Traditionally, we find two kinds of criterion-based validity: concurrent and predictive (Bachman 1990, 248). Concurrent criterion relatedness involves one of the several commonly employed procedures: examining differences in test performance among individuals at different levels of proficiency, examining correlations among different measurements of the given ability, correlations with existing standardized tests, comparisons with teacher ratings, with informant data, with self-assessment, and with different versions of the same test (Weir 2005, 209). Predictive criterion relatedness focuses on demonstrating a link between test scores and some future performance, where the test scores predict the criterion behavior of interest — for example, having a writing test serve as a predictor of place allocations in writing courses of different levels, or linking predictive criterion validity to comparison with external performance benchmarks such as the CEFR, for example (Weir 2005, 209). In particular, for the purposes of writing assessment, Shaw and Wier list the following three procedures (2008, 230):
•	cross-test comparability: the procedure encompasses the effort of comparing different and yet related language proficiency measures (such as comparing the ratings of two different tests and ascertaining the levels of their relatedness);
•	comparison with different versions of the test: relies on the existence of a relationship between two or more different versions of the test involving the same test takers and other comparable conditions; and
•	comparison with external standards: using standardized and reputable existing measurements (such as national graduation (Matura) exams or well-known British Council or ESOL examinations) as benchmarks for making comparisons to.
The statistical methodology of pursuing further aspects of a posteriori validity then becomes increasingly more complex than just looking at the correlation coefficients one would normally find in criterion-related validity. This is the reason why we mostly see them in more serious
18 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E L O P
validation efforts, starting with the focus on scoring validity in an empirical sense. A posteriori analysis of the obtained scores looks into rating scales, inter- and intra-rater agreement, moderation of scores, and then beyond into use and consequences (Weir 2005), and as such involves complex dealing with quantitative data. A posteriori analysis of scoring validity involves focusing on two major aspects (Shaw and Weir 2008, 190) that are also strongly grounded in statistical methodology:
•	post-exam adjustments: they include statistical methods (such as scaling or Rasch measurements) aimed at artificially correcting scores based on the attested interplay of different features surrounding the test (severity and leniency, central tendency, and more); and
•	grading and awarding: the final process is the one of assigning cut-off scores and issuing certificates (if applicable).
However, one can argue that this part of empirical validation deals much more with consistency of measurement (and hence reliability) than what we would normally understand as validity, even if we look at them as merged. Hence, most a posteriori efforts at validation in the strictest sense revolve around construct validation in its narrow meaning. Empirical construct validation focuses on finding empirical evidence (in terms of correlations and experimentation) which goes towards confirming or disproving a particular interpretation of the obtained test scores (Bachman 1990, 260). It functions in terms of two very broad approaches at gathering validity evidence — correlational (and other psychometric methods) and experimental approaches. The correlational approaches can be seen as either exploratory or a confirmatory. The exploratory approach concentrates on identifying the traits that influence test performance by examining the correlations4 among a large set of measures. The confirmatory approach focuses on a particular hypothesis about the traits and attempts to confirm or reject it. On the other hand, the classical experimental design for testing hypotheses (which has up to recently received less attention) involves attempts at manipulating the variables involved in a testing situation, such as example pre-test/post-test experimentation, and similar (Bachman 1990, 258-66).
Finally, looking beyond the test and the scores, we have the social effect a test use may have, usually observed within the framework of consequential validity. This involves us considering the ethical basis of validity as incorporating deliberations which are neither scientific nor technical, and which focus on the influence of a particular (educational) system on the interpretation of a test, as well as on the washback effect that test use has on that particular system in reverse (1990, 279). In practice, this means observing (see Bachman (1990, 280-84) and Weir (2005, 210)):
the rights of the test-takers (secrecy, confidentiality, privacy); the values inherent in test developers and raters; the values inherent in the particular social system; background knowledge; and influence on teaching and learning.
Related psychometric approaches also include factor analysis, multi-trait-multi-method investigation, and more.
MME1	LANGUAGE 19
5 Direct Assessment of Writing - The Cutting Edge and Outlook
If we review the entire discussion presented in the previous two sections, looking at both the general push and pull within development of writing assessment and at the issues of reliability and validity within language testing in general, and assessing writing in particular, we can with some degree of confidence suggest a global outline of what validity in terms of direct assessment of writing entails, and how it can be supported:
•	validity entails the application of scientific rigor to the process of interpretation of test scores by focusing on both rational argument and empirical evidence;
•	even though it can be observed as a conceptual whole, validity is in essence comprised of different aspects and this division is extremely useful when it comes to the practical (empirical) conduct of validation efforts;
•	reliability is to be considered as an integral part of overall validity, and especially so when it comes to writing assessment;
•	writing as a language skill deserves special consideration when it comes to its theoretical description within the larger concept of language competence; and
•	equal attention should be given to the a priori and a posteriori efforts of ensuring validity.
If we were to translate this general overview, emerging from the overall discussion, into practical steps needed to assure the highest possible degree of validity, we can dilute them into a validation checklist, as follows in Figure 1 below (largely revolving around and adapted from the Shaw and Wier (2007)). Following a list such as this one can lead any practitioner to easily pinpoint most of the steps they need to take when setting up the assessment, and can/should take after they have administered the test, in order to ensure the highest level of validity in the context of the notoriously problematic phenomenon that is the assessment of writing.
In the end, when it comes to further development of securing validity within assessing writing ability, it is hard to foresee any new theoretical movements in the future. The core deliberations regarding the nature of validity in educational (and psychological) testing took place between 1950s and 1990s, culminating in the extensive account presented by Mesick in 1989. Similarly, Bachman's (1990) translation of Messick's (1989) influential account into the framework of language testing basically set the ultimate theoretical underpinnings of validity for this particular field, and even added the somewhat novel consideration of incorporating reliability within validity. However, Bachman's account cannot be seen as bringing anything conceptually new with regards to Messick (perhaps only adding the said notion of reliability being merged with validity), nor can any other account seen as coming after his. The real contribution he and other of the authors discussed in this paper made was in furthering the practice of validation and fine-tuning the outline of the sources of evidence and procedures one should focus on to address different aspects of unified validity. This is the direction the development of validity and validation research in terms of language assessment, and in particular that of writing, will most likely follow in the future, with conceptual novelty highly unlikely in this content (apart from any possible changes related to our understanding of the construct itself).
One of the areas that will see particular development is the validation of scoring, both in an a priori and a posteriori sense. A priori, the development will most likely focus on rating scales and
20 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E
L
□
P E
Figure 1. A summaryof the prototypicalvalidity effortofanywriting assessment.
their application. For instance, one of the more interesting directions the advance of rating scales is taking is in moving away from the vague 'can do' and 'has got' descriptors often found in rating scales to the actual (linguistic) features being weighted and ticked off as present (or not) in relation to ratings. This represents the first step of actually turning away from a purely 'judging' perspective predominant in rating writing performances to a more 'counting' oriented one. Likewise, a posteriori, the shift to a more 'counting' oriented approach would mean the introduction and reliance on new methods of score adjustment, quality assurance (via agreement studies), computer-aided processes being introduced on a wider scale, eye-tracking investigations, and more.
MME1	LANGUAGE 
21
References
Alderson, J. Charles, Caroline Clapham, and Dianne Wall. 1995. Language Test Construction and Evaluation. Cambridge: Cambridge University Press.
American Psychological Association. 1955. Standards for Educational and Psychological Testing and Manuals. Washington, DC: Author.
Applebee, Arthur N. 1981. Writing in the Secondary School: English and the Content Areas. Urbana, IL: National Council of Teachers of English.
Bachman, Lyle F. 1990. Fundamental Considerations in Language Testing. Oxford: Oxford University Press.
Bachman, Lyle F., and Adrian S. Palmer. 1996. Language Testing in Practice: Designing and Developing Usful Language Tests. Oxford: Oxford University Press.
Breland, Hunter M. 1983. The Direct Assessment ofWriting Skill: A Measurement Review. College Board Report No. 83—6. New York: Educational Testing Service.
Brown, Rexford. 1978. "What We Know Now and How We Could Know. More about Writing Ability in America." Journal of Basic Writing 1 (4): 1—6.
Campbell, David, and D. Fiske. 1959. "Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix." Psychological Bulletin 56 (2): 81-105. https://doi.org/10.1037/h0046016.
Campbell, David, and Julian C. Stanley. 1966. Experimental and Quasi-Experimental Designs for Research. Boston: Cengage Learning.
Chan, Eric K. H. 2014. "Standards and Guidelines for Validation Practices: Development and Evaluation of Measurement Instruments." In Validity and Validation in Social, Behavioral, and Health Sciences, edited by Bruno D. Zumbo and Eric K. H. Chan, 9-24. New York: Springer.
Coffman, William E. 1971. "On the Reliability of Ratings of Essay Examinations in English." Research in the Teaching of English 5: 24-36.
Cooper, Charles R. 1977. "Holistic Evaluation of Writing." In Evaluating Writing: Describing, Measuring, Judging, edited by Charles R. Cooper and Lee Odell, 3-31. Urbana: National Council of Teachers of English.
Cooper, Charles R., and Lee Odell, eds. 1977. Evaluating Writing: Describing, Measuring, Judging. Urbana: National Council of Teachers of English.
Coward, Ann F. 1952. "A Comparison of Two Methods of Grading English Compositions." Journal of Educational Research 46 (2): 81-93. https://doi.org/10.1080/00220671.1952.10882003.
Cronbach, Lee J. 1990. Essentials of Psychological Testing. 5th Ed. New York: Harper & Row.
Cronbach, Lee J., and Paul E. Meehl. 1955. "Construct Validity in Psychological Tests." Psychological Bulletin 52: 281-302. https://doi.org/10.1037/h0040957.
Davies, Alan, ed. 1968. Language Testing Symposium. A Psycholinguistic Perspective. London: Oxford University Press.
Davies, Alan, and Catherine Elder. 2005. "Validity and Validation in Language Testing." In Handbook of Research in Second Language Teaching and Learning, edited by Eli Hinkel, 795-813. Mahwah, New Jersey: Lawrence Erlbaum Associates.
D'Este, Claudia. 2012. "New Views ofValidity in Language Testing." EL.LE 1 (1): 61-76. https://doi. org/10.14277/2280-6792/5p.
Diederich, Paul B., John Winslow French, and Sydell T. Carlton. 1961. "Factors in Judgments of Writing Ability." ETS Research Bulletin 1961 (15): i-93. https://doi.org/10.1002/j.2333-8504.1961.tb00286.x.
Diederich, Paul B. 1974. Measuring Growth in English. Urbana: National Council of Teachers of English.
22 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
Douglas, Dan. 2000. Assessing Language for Specific Purposes. Cambridge: Cambridge University Press.
Ericsson, K. Anders, and Herbert A. Simon. 1993. Protocol Analysis: Verbal Reports as Data. Cambridge, MA: MIT Press.
Godshalk, Fred I., Frances Swineford, and William Eugene Coffman. 1966. The Measurement of Writing Ability. New York: College Entrance Examination Board.
Hobbs, Catherine L., and James A. Berlin. 2001. "A Century of Writing Instruction in School and College
English." In A Short History of Writing Instruction: From Ancient Greece to Modern America, edited by James J. Murphy, 247—89. Mahwah: Lawrence Erlbaum Associates.
Hopkins, Levi Thomas. 1921. The Marking System of the College Entrance Examination Board. Cambridge: The Graduate School of Education, Harvard University.
Huddleston, Edith M. 1952. "Measurement of Writing Ability at the College-Entrance Level: Objective vs. Subjective Testing Techniques." ETS Research Bulletin Series 1952 (2): 165—213. https://doi. org/10.1002/j.2333-8504.1952.tb00925.x.
Hunt, Kellogg W. 1977. "Early Blooming and Late Blooming Syntactic Structures." In Evaluating Writing:
Describing, Measuring, Judging, edited by Charles Cooper and Lee Odell, 91—106. Urbana: National Council of Teachers of English.
Huot, Brian. 1990. "Reliability, Validity, and Holistic Scoring: What We Know and What We Need to Know." College Composition and Communication 41 (2): 201—13.
Huot, Brian, Peggy O'Neill, and Cindy Moore. 2010. "A Usable Past for Writing Assessment." College English 72 (5): 495-517.
Jacobs, Holly L., Stephen A. Zinkgraf, Deanna R. Wormuth, V. Faye Hartfiel, and Jane B. Hughey. 1981. Testing ESL Composition: A Practical Approach. Rowley, MA: Newbury House.
Kane, Michael T. 1992. "An Argument-Based Approach to Validity." Psychological Bulletin 112 (3): 527-35. https://doi.org/10.1037/0033-2909.1123.527.
—. 2001. "Current Concerns in Validity Theory." Journal of Educational Measurement 38 (4): 319-42. https:// doi.org/10.1111/j.1745-3984.2001.tb01130.x.
Lado, Robert. 1961. Language Testing: The Construction and Use of Foreign Language Tests: A Teacher's Book. New York: McGraw-Hill.
Loevinger, Jane. 1957. "Objective Tests as Instruments of Psychological Theory." Psychological Reports 1957 (3); 635-94. https://doi.org/10.2466%2Fpr0.1957.33.635.
McIntyre, Philip N. 1993. "The Importance and Effectiveness of Moderation Training on the Reliability of
Teachers' Assessments of ESL Writing Samples." MA Thesis, Department of Applied Linguistics, University of Melbourne.
McNamara, Tim F. 1996. Measuring Second Language Performance. Harlow: Addison Wesley Longman.
Messick, Samuel. 1975. "The Standard Problem. Meaning and Values in Measurement and Evaluation."
American Psychologist 30 (10): 955-66.
—. 1989. "Validity." In Educational Measurement, edited by Robert Linn, 13-103. Washington, DC: American Council on Education and National Council on Measurement in Education.
Miller, David M., and Robert L. Linn. 2000. "Validation of Performance-Based Assessments." Language Testing 24 (4): 367-78. https://doi.org/10.1177%2F01466210022031813
Mullis, Ina V. S. 1980. Using the Primary Trait System for Evaluating Writing (Report No. 10-W-51). Denver: National Assessment of Educational Progress, Education Commission of the States.
Nikolov, Marianne, ed. 2009. Early Learning of Modern Foreign Languages: Processes and Outcomes. Clevedon: Multilingual Matters.
MME1	LANGUAGE 23
Ruch, Giles Murrel. 1929. The Objective or New-Type Examination: An Introduction to Educational Measurement. Chicago: Scott, Foresman and Co.
Scott, W. 1917. "A Fourth Method of Checking Results in Vocational Selection." Journal of Applied Psychology 1: 61-66. https://doi.org/10.1037/h0073494.
Shaw, Stuart D., and Cyril J. Weir. 2007. Examining Writing: Research and Practice in Assessing Second Language Writing. Cambridge: Cambridge University Press.
Starch, Daniel, and Edward C. Elliott. 1912. "Reliability of the Grading of High-School Work in English." School Review 20 (7): 442-57.
Sheppard, Everett M. 1929. "The Effect of Quality of Penmanship on Grades." Journal of Educational Research 19 (2): 102-5.
Sigott, Günter. 1994. "Language Test Validity: An Overview and Appraisal." AAA: Arbeiten aus Anglistik und Amerikanistik 19 (2): 287-94.
—. 2004. Towards Identifying the C-Test Construct. Frankfurt am Main: Peter Lang.
Stalnaker, John M. 1936. "The Problem of the English Examination." Educational Record 17: 140-43.
Thorndike, Edward L. 1918. "The Nature, Purposes and General Methods of Measurements of Educational
Products." In The Measurement of Educational Products - National Society for the Study of Education Yearbook, edited by Guy Monteose Whipple, 16-24. Chicago: National Society for the Study of Education.
Traxler, Arthur E., and Harold A. Anderson. 1935. "Reliability of an Essay Test in English." School Review 43 (7): 534-40.
Tsai, Constance Hui Ling. 2002. "Issues of Validity in the Assessment of Writing Performance." Teachers College, Columbia University Working Papers in TESOL & Applied Linguistics 4 (2): 1-3.
Tyler, Ralph W. 1934. Constructing Achievement Tests. Columbus: Bureau of Educational Research, Ohio State University.
Weigle, Sara C. 1994. "Effects of Training on Raters of ESL Compositions." Language Testing 11 (2): 197-223. https://doi.org/10.1177%2F026553229401100206.
—. 2002: Assessing Writing. Cambridge, UK: Cambridge University Press.
Weir, Cyril J. 2005. Language Testing and Validation: An Evidence-Based Approach. Houndgrave, UK: Palgrave.
Weir, Cyril J., Ivana Vidakovic, and Evelina D. Galaczi. 2013. Measured Constructs: A History of Cambridge English Language Examinations 1913—2012. Cambridge: UCLES/Cambridge University Press.
White, Edward M. 2001. "The Opening of the Modern Era of Writing Assessment: A Narrative." College English 63 (3): 306-20. https://doi.org/10.2307/378995.
Wiseman, Stephen. 1949. "The Marking of English Compositions in Grammar School Selection." British Journal of Educational Psychology 19: 200-209. https://doi.org/10.1111/j.2044-8279.1949.tb01622.x.
—. 1956. "Symposium: The Use of Essays in Selection at 11+." British Journal of Educational Psychology 26: 172-79. https://doi.org/10.1111/j.2044-8279.1957.tb01390.x.
Yancey, Kathleen Blake. 1999. "Looking Back as We Look Forward: Historicizing Writing Assessment." College Composition and Communication 50 (3): 483-503. https://doi.org/10.2307/358862.
Zumbo, Bruno D. 2007. "Validity: Foundational Issues and Statistical Methodology." In Handbook of Statistics 26: Psychometrics, edited by C. R Rao and S S. Sinharay, 45-79. Amsterdam: Elsevier Science.
—. 2009. "Validity as Contextualized and Pragmatic Explanation, and Its Implications for Validation Practice."
In The Concept of Validity: Revisions, New Directions and Applications, edited by Robert W. Lissitz, 65-82. Charlotte: Information Age Publishing.
24 Nikola Dobric Reliability, Validity, and Writing Assessment: A Timeline
E L O P
Vesna LazoviC	2018' VoL 15 (2), 25-44/(!20)
TT . . ri. ...	.	revije.ff.uni-lj.si/elope
University of Lpbljam, Slovenia	doi: io.4312/elope.15.2.25-44
UDC: [811.111'33:811.163.41]:[336.71:004.738.5:659.1]
How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and Serbian in Light of Relevance Theory
ABSTRACT
Based on a corpus comprising online advertisements of banks operating in the UK and Serbia, this paper aims at describing the most frequent ways of creating puns and classifying them based on their predominant structure. The function of puns in advertising is explained from the perspective of Sperber and Wilson's relevance theory (1995) and the behavioural AIDA model. This paper also tries to reveal whether the formal and semantic aspects of lexical units are exploited in British and Serbian bank offers, and whether there are any language- and culture-specific features in terms of punning.
Keywords: puns; bank advertisements; relevance theory; English; Serbian
Kako se igrati/plačati z besedami? Raba besednih iger v spletnih bančnih oglasih v angleščini in srbščini v luči teorije relevantnosti
POVZETEK
Na podlagi analize korpusa spletnih bančnih oglasov v Britaniji in Srbiji v članku obravnavamo najpogostejše načine tvorjenja besednih iger ter njihovo klasifikacijo glede na zgradbo. Vlogo besednih iger v oglaševanju razložimo z vidika teorije relevantnosti (Sperber in Wilson 1995) ter behaviorističnega modela AIDA. V članku tudi poskušamo ugotoviti, ali so v britanskih in srbskih bančnih ponudbah izkoriščeni formalni in semantični vidiki leksikalnih enot, kakor tudi obstoj jezikovnih in kulturnih značilnosti tvorjenja besednih iger.
Ključne besede: besedne igre; bančni oglasi; teorija relevantnosti; angleščina; srbščina
MME1
LANGUAGE 25
How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and Serbian in Light of Relevance Theory
1	Introduction
Advertisements are very often characterised by the inventive use of language and unorthodox semantic functions of words. Figures of speech, also known as tropes and rhetorical figures, are widely employed, especially, as Leech (1966, 183) points out, because of their striking and memorable quality. Dyer (1982, 152) states that the purpose of figurative language is to create effects, attract attention and arouse emotions by breaking or exploiting rules. In this way, she argues, emotive associations are linked to the product.
A pun as a rhetorical figure plays an important role within the persuasion process, since it cleverly disguises certain messages in advertisements. McGuire (2000) emphasises the persuasion effect of using figurative language in communications and claims that rhetorical figures are often used: (a) to enhance attention to an advertisement; (b) to increase its believability along with the amount of information a consumer remembers; (c) to influence a consumer's mood; and finally, (d) to persuade them to purchase products or services. Sternkopf (2005, 227) adds that advertising language is highly ambiguous, since clear and concrete statements are avoided and vague utterances are used to activate numerous layers of potential interpretation.
This paper aims to explain the frequent puns1 employed in bank advertisements and to highlight their structure. It is interesting to observe how language is used in financial advertising, since people are generally cautious about the claims and tempting messages in bank offers. The idea is to explore whether puns are similarly used in two languages, and whether there are any culture-specific features attached to their use.
2	Theoretical Framework
The function of puns in advertising is explained and analysed in this paper from the perspective of Sperber and Wilson's relevance theory (1995), an inferential approach to pragmatics, later slightly modified and updated (Wilson and Sperber 2004), and the behavioural AIDA model, both of which are explained in the following sections.
2.1 Relevance Theory
Building upon Grice's cooperative principle and violation of maxims (1989), Wilson and Sperber centre their argument around the claim that the expectations of relevance raised by an utterance are precise and predictable enough to help the hearer infer the speaker's meaning (2004, 607). According to relevance theory, the search for relevance is a basic feature of human cognition. Hence, any input (e.g. an utterance) is regarded as relevant "when its processing in a context of available assumptions yields a positive cognitive effect" (Wilson and Sperber 2004, 608). It is argued that the more positive cognitive effects an utterance produces and the less processing effort it requires, the more relevant that utterance will be.
In this paper both terms — puns and wordplays — will be used interchangeably as synonyms to indicate the same deliberate semantic exploitation of words and phrases.
26 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
The concept of ostensive-inferential communication is crucial in this theory. There are two intentions in communication:
a.	the informative intention (the intention to inform the hearer of something) and
b.	the communicative intention (the intention to have one's informative intention recognised)
Once the speaker's informative intention is recognised, the communicative intention is fulfilled and understanding is achieved. In other words, the hearer must first realise that the speaker has the intention to communicate something. The process of this communication actually involves the use of an ostensive stimulus "designed to attract an audience's attention and focus it on the communicator's meaning" (Wilson and Sperber 2004, 611), which means that only relevant stimuli will be noticed. More precisely, this stimulus needs to be optimally relevant, i.e. to fulfil the following two conditions if communication is to be successful: (1) to be relevant enough to be worth the audience's processing effort; and (2) to be the most relevant and economical one for the audience to process and understand it easily. Otherwise, the utterance will be uninterpretable and the speaker's meaning unidentifiable.
Tanaka (1994, 26) points out that although an utterance may communicate many assumptions, not all of them are intended at the same time by the speaker, as some are purely accidental. Among the intended assumptions, some are communicated explicitly while others are implicit. According to Sperber and Wilson (1995, 182), explicatures are assumptions obtained by the development of the logical forms encoded by an utterance. On the other hand, implicatures include contextual assumptions (implicated premises) and contextual implications (implicated conclusions) derivable from the proposition expressed by the utterance (1995, 194—195). They underline the fact that comprehension process is not linear. Instead, "hypotheses about explicatures and implicatures are developed in parallel against a background of expectations which may be revised or elaborated as the utterance unfolds" (Wilson and Sperber 2004, 615). The hearer starts with the decoded linguistic meaning with the least processing effort, which must be then contextually enriched at the explicit level and complemented at the implicit level up until the resulting interpretation meets his expectation of relevance and the speaker's meaning is yielded (Wilson and Sperber 2004, 613). Along the way there may be ambiguities to resolve, ellipses to interpret, metaphors and ironies to interpret, etc.
In terms of online advertising, attracting attention seems to be rather a challenging task because online competition is so vast and seemingly unlimited. Users now have greater control over choosing sites and products they see online or even excluding the advertising content with the help of application software. Therefore, many advertisements can be unnoticed or completely ignored. In other words, if the ostensive stimulus produced by the addresser fails to catch the attention of the audience, no communication can take place. Alternatively, when a potential buyer or client pays attention to a given ostensive stimulus, they find it relevant and want to continue processing it.
Byrne (1992) uses relevance theory for the analysis of the text to illustrate the role of linguistic style in advertising, which is used as a tool for conveying meanings beyond the ones verbalised. Such analysis can thus help in disambiguation, examination of different contextual implications and some typical features of advertising text. Tanaka (1994, 64) thinks that relevance theory offers the best framework for the analysis of the role of puns in advertising and the question of processing effort. Puns, however, may present a problem for this approach, because extra processing effort is needed for their interpretation. In this case, puns can be analysed as cases
MME1	LANGUAGE 27
of layering in communication. As Tanaka (1992, 96) argues, puns do achieve optimal relevance exactly because the ostensive stimulus is the most economical way for the advertiser to achieve many of the intended contextual effects. This means that the production of an utterance which is apparently uninterpretable at one level may be used as an ostensive stimulus at another (Sperber and Wilson 1987, 751). Without going through the whole process, the audience would ignore the advertising caption and the message would lose its strength.
To put it in simpler terms, puns are extensively used in advertising although they make the hearer invest additional processing effort which would influence relevance. Nonetheless, positive cognitive effects compensate for the extra processing efforts used to resolve the pun, and these outweigh potential disadvantages (Díaz-Pérez 2012, 29). However, if a pun requires too much processing effort, the effect may be counterproductive, as it can produce confusion or even scepticism towards the product.
From the perspective of relevance theory, puns in advertising can be analysed in terms of the interpretation process involved (based on Tanaka 1994, 62—82)2 in the following ways:
1.	two meanings are activated; one interpretation is to be rejected since it makes no sense and gives place to the intended one — nonsense puns,
2.	the rejected interpretation provides access to encyclopaedic knowledge, which is then used in processing the intended interpretation — context-based puns,
3.	one interpretation is activated; however, the message insinuates some things otherwise indecent or inappropriate to be overtly expressed — insinuating puns,
4.	two (or more) meanings are activated and both of them are actually communicated — complex puns.
Tanaka (1994, 62) states that "the essence of the pun lies in its access to multiple interpretations". Puns are successful if the hearer activates more than one interpretation of a given utterance no matter whether one interpretation is intended to be rejected or not. If two meanings are to be recovered and communicated, they have to reinforce one another.
Whatever linguistic or extralinguistic devices are used, the advertiser's main intentions are always the same — attracting and maintaining the audience's attention, making them remember the advertisement, associating the product with desirable qualities and persuading them to buy it. This is directly related to the AIDA model discussed in the next section.
2.2 AIDA Model
In order to properly understand the reasons for employing wordplay in advertising, we also need to emphasise a cognitive marketing concept crucial for the main aspects and functions of advertising communication, i.e. the behavioural AIDA model developed by Elmo Lewis (Strong 1925). This acronym stands for (attracting) Attention, (raising) Interest, (establishing) Desire and (initiating) Action, describing the sequential mental processes that may occur when the consumer sees an advert. It is used as a basic framework to explain how an advertising message is created, how it works and what linguistic means are used to make it effective. Different functions are realised with the use of language forms and images, i.e. different means can be used for
2 The boundaries between these types of pun are not clearcut, but rather a matter of degree (Tanaka 1994, 65). The ultimate message is reached when more than one interpretation is combined.
28 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
a specific function (e.g. the use of trigger words, the choice of syntactic forms, direct reader addressing, metaphors, puns, etc.).
According to Kotler and Armstrong (2012, 417-18), the AIDA model suggests the desirable qualities of a good and effective message, which should, ideally, get attention, hold interest, arouse desire, and obtain action. However, not many messages actually take the consumer through all four steps, from awareness to purchase; instead, the model presents a solid theoretical starting point for establishing advertising objectives. They, further, emphasise that marketing communicators must decide what to say (the message content) and how to say it (the message structure and format).
Regarding the message content, the communicator first identifies an appeal that will produce a desired response and motivate purchase. There are three types of appeals:
(a)	rational appeals,
which relate to the audience's self-interest and show the product's benefits;
(b)	emotional appeals,
which provoke either positive (love, joy, humour) or negative emotions (fear, guilt);
(c)	moral appeals,
which are directed to an audience's sense of what is right or wrong.
Regarding the structure and format, communicators have a range of different options to use. Structurally, they can, for example, draw conclusions for the product or let buyers make their own inferences by asking questions; they can present the strongest arguments first or last; or they can mention only the product's strengths or also admit its shortcomings. Formally, they can attract attention by using novelty, eye-catching pictures and headlines, distinctive colours, size or position, and so on.
Finally, Horky (2009) suggests how the AIDA model can be used in internet marketing, which is shown in Diagram 1. He notices that business objectives can be reached by combining the characteristics of traditional and internet marketing with elements of effective marketing communication models.
As can be seen from the diagram, there are numerous ways to realise the AIDA steps. However, it is quite rare to find an advert with a one-to-one relationship with the various steps proposed by the concept, since two or more steps normally merge together (Vestergaard and Schroder 1985, 50). Consequently, the functions of language partially overlap and coincide, so a clear demarcation is not always possible.
The use of puns, as will be shown throughout the paper, is an effective way to realise all these steps, and thus they are frequently applied in commercial advertising. Puns surprise and provoke readers, as they have to decipher deeper meanings and solve a puzzle, after which they become interested in the product and they are more likely to eventually buy it. The product is hence promoted by "creating humour, attracting the reader's attention and adding persuasive force to the message" (Laviosa 2005, 25). With a puzzle to solve, the reader's attention is captured and the ultimate message, be it rather simple and obvious, is much more appealing.
MME1	LANGUAGE 29
• unexpected content, situation, animation
•	shocking facts or statistical data
■ thought-provoking questions
• surprise
•	attractive graphics
or titles
■	relevant message^,
• illustrations
■	promise of reward
or satisfaction
• raising tension or mystery
■ personalization J

J
•	communicating unique benefits
• providing a solution for the problem
•	building unique brand image and
V.^must-to-have effect J
• special offers
•	sense of urgency
•	immediate actions (purchase, order,
subscription)
•	sending a message through an online
form__J
Diagram i . AIDA model in internet marketing.
3 Puns in Advertising
Puns refer to "the use of words or phrases to exploit ambiguities and innuendoes in their meaning, usually for humorous effect" (Collins Cobuild online dictionary). More precisely, as Delabastita (1996, 128) explains, puns contrast "linguistic structures with different meanings on the basis of their formal similarity", which can be complete — homonymy, or partial — homophony, homography and paronymy. The context itself is vital, since it allows for multiple readings to be triggered and the pun to be activated and realised. Otherwise, these instances are regarded as unintentional wordplay. As Partington (2009, 1799) explains, "it is the interplay, the enforced switching from one mode of interpretation to another [...] which is at the heart of a great part of wordplay."
Crystal (2003, 405) points out that the two unrelated meanings of a single word are suddenly and unexpectedly brought together, and the consequent incongruity makes us laugh. Apart from their humorous effect, wordplay also captures and holds the attention of the audience (for different functions of wordplay, cf. Thaler 2016, 51-52). This is a very useful attention-getting device based on intended ambiguity and alternative interpretations, which arise from the confrontation of similar forms but different meanings. According to Leech (1966, 184), ambiguity can be defined as a many-to-one relationship between levels, where:
a)	different meanings are expressed by using the same word — polysemy, as in one British bank offer:
Sort your overdraft
where the verb to sort triggers two possible meanings — 'to arrange' and 'to deal with'; or
b)	different meanings are expressed by words with the same phonological or graphological form — homonymy, as in one Serbian bank offer:
30 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
Nije u redu da cekate u redu. Predite na e-banking!
[It is not all right to wait in a queue. Switch to e-banking!]3
where the Serbian phrase u redu is repeated in two different senses, as a phrase — 'to be all
right / ok' and as a preposition + noun — 'in a queue'.
Even though they are thought to be inappropriate for serious discourse, due to their humorous nature and intention, puns are a perfect choice for advertising (Sherzer 1985). Redfern (1984, 131) thinks that the utility of puns in this context lies in the fact that they are highly economical in terms of advertising space, otherwise rather costly.
Although both advertising agencies and some influential copywriters in the 1960s and 70s (e.g. Ogilvy 1964; Caples 1974) were reluctant to admit that using a pun is extremely productive and effective, many research studies that followed proved that puns are popular rhetorical figures in advertisements. For instance, Leech (1966) lists the inventive and imaginative use of language as an important linguistic device available to copywriters when creating advertising material; Redfern (1984) thoroughly analyses puns using examples from advertisements and literature; Cook (1992) discusses the impact of puns in perfume and car advertisements; Tanaka (1994) observes that punning is an essential device in advertising; Myers (1994) speculates that puns appeal to readers because more work and time is required to process them and explains some popular examples; Leigh (1994) finds an abundance of puns in print ad headlines; both McQuarrie and Mick (1996) and Tom and Eves (1999) highlight the importance of language games and ambiguity. In an experimental setting, van Mulken, van Enschot-van Dijk, and Hoeken (2005) show that the presence or absence of puns has a significant impact on the respondents' appreciation of the related slogans. They state that puns give the audience both a pleasurable experience and intellectual satisfaction after solving a semantic riddle, which in turn may lead to a more positive attitude towards the product (van Mulken, van Enschot-van Dijk, and Hoeken 2005, 708). Furthermore, Djafarova (2008) explores the interpretation processes of puns in print advertisements of tour operators and discusses intended meanings in relation to the background and inferential knowledge of the audience. Labrador et al. (2014) focus on the rhetorical structure and the linguistic features of persuasive language in online advertisements of electronic products. They also identify puns and regard them as instances of language play indicating informal style. In a marketing-oriented research study, Choi, Taylor, and Lee (2017) investigate whether advertisements with puns are more effective and if they influence consumers' attitude toward the brand.
Although numerous research studies have been conducted with the aim of revealing semantic peculiarities in commercial advertising, very few have addressed and investigated the matter in a specific advertising sector — financial products and services (cf. Albers-Miller and Straughan 2000; Adams and Cruz Garcia 2007; Lazovic 2014). Trust in banks has declined in recent years due to the significant, negative impact the global financial crisis has had on ordinary people's lives. Moreover, as a result of severe competition, banks are constantly trying to restore and retain trust by offering safety, security and reliability in the post-crisis period. For that reason, advertisements for financial products have their own unique features compared to other types of advertisements, such as the abundance of restrictions, the key role of numbers, factual and objective tone, infrequent use of neologisms, etc. (Adams and cruz Garcia 2007). Bank advertisements present banks as trustworthy and reliable, while emphasising their stability and security. They often play
3 All the examples in Serbian are translated into English by the author.
MME1	LANGUAGE 31
on audiences' aspirations, which, as they promise, are easily affordable and accessible. They are regarded as more serious compared to advertisements for cosmetic products, food and drink products, etc. However, they also share some typical or universal characteristics, such as short, dense texts, usually with the abundance of ellipsis, in the form of slogans, clichés and telegraphic descriptions of a product emphasising its advantages (Lazovic 2014).
In light of relevance theory, advertisements use a range of devices to attract the attention of potential buyers. These devices frustrate the expectation of relevance they have created, because the audience will try to discover the hidden meaning. The ultimate message is usually obvious, although disguised so as to be more appealing and challenging to decipher. Puns, for example, create extra positive cognitive effects, and in such a way increase the relevance of both the advertisement and the message(s) conveyed. The message is in this way communicated with more strength. While solving the puzzle the audience's attention is retained longer, and thus once the pun is comprehended it is often remembered (Tanaka 1994, 71). Increased memorability is thus a major advantage derived from more processing effort. The longer the pun takes to process, the more likely the audience will memorise it. On the other hand, if there is no pun and the message is straightforward, this considerably lowers the chances of the audience remembering advertisements and possibly reacting to them.
Surprisingly enough, the advertiser is not concerned with a positive impact of the pun. Even if some people find a pun obscure or irritating, this will still be welcomed by the advertiser, since they consider that in terms of product recognition evoking any kind of reaction, either positive or negative, is better than evoking no reaction.
4 Research Questions and Research Methodology
This research study investigates puns in advertisements found in the UK and Serbian bank offers online. The analysis of online bank advertisements is conducted in order to reveal the structure of puns exploited in the two languages. Several research questions are addressed:
1.	Are puns a common technique in the UK and Serbian bank advertisements?
2.	Is the structure of puns the same in the two languages?
3.	Does the usage vary between languages and is this variation culturally-bound?
4.	What are possible interpretations of puns in light of relevance theory?
This comparative analysis features a particular type of advertisements, namely bank advertisements, because advertisements for financial services are thought to be among the most boring and predictable of those from any sector (Myers 1999, 38). Jewler (1992, 49) notes the problems copywriters face when they have to write a caption for a bank account and struggle to think of yet another way of saying that "our interest rates on long-term certificates of deposit aren't the worst in town". In addition, it is a highly competitive category, with new banks constantly emerging. It is thus very challenging for banks to compete, because in the public's mind there is not much that differentiates one institution from another (Jewler 1992, 49). Finally, bank advertisements are chosen in order to examine the claim proposed by Adams and Cruz García (2007, 143) that press adverts for financial products infrequently use puns, and to investigate whether this applies to online adverts as well.
The analysis of different puns in this paper is based on the self-compiled corpus. Online bank advertisements were continually followed and collected over the course of one year, i.e. from
32 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
May 2012 to May 2013. The analysed corpus comprises 570 Serbian and 770 UK online bank advertisements offering services on the official home pages of the 32 most prominent banks operating in the UK and 33 banks operating in Serbia at the time of corpus collection. All the advertisements are planned, aimed and directed at consumers rather than corporations. In order to provide a consistent and uniform data sample of advertising text to be analysed, the following criteria were applied for advertisement selection:
1.	An advertisement is found on the homepage of the commercial bank regardless of the position on the page and the size.
2.	An advertisement either features bank's services or services of their partners.
3.	An advertisement for financial products is directed at the general public rather than at experts.
4.	An advertisement is either text-based (e.g. only slogans), graphic-based (e.g. only images) or most often, both text- and graphic-based.
5.	Multimedia elements, such as audio and video sequences, are not included in the corpus.
6.	The product or service advertised belongs to the retail personal banking sector, not private, business or corporate banking sector.
7.	Only the text found in the advertisement is analysed, whilst detailed descriptions and explanations4 found on the transferred sites are excluded.
For this type of comparative analysis, the instances of puns are first identified and then classified according to their underlying structure. The representative examples are further interpreted within the advertising context following the principles of relevance theory to reveal hidden meanings. The study is qualitative and exploratory in nature. Since the discussion centres around the verbal component, visual puns are excluded from any further discussion.
5 Findings
This part focuses on different structures of puns, examples of which are illustrated and explained. The primary purpose of the present study is to build upon the prior research (Tanaka 1992, 1994; Byrne 1992; Díaz-Pérez 2000, 2012) and examine the use of puns in bank advertisements using the two frameworks previously introduced.
In order to make puns effective, different linguistic features are exploited along with different forms of knowledge about the world and culture. Thaler (2016, 52) observes that there is an almost unlimited range of possibilities to modify linguistic material creatively, which makes it difficult to offer any comprehensive list of wordplay structure (for one possible taxonomy of puns in English cf. Koren 2012).
Based on the recurrent types of wordplay in bank adverts and different language levels exploited, the classification offered here is a slightly modified and expanded version of the one found in Delabastita (1996, 130—31). The idea is to illustrate and describe the most common instances found in both corpora. Puns are therefore classified into six groups, five of which concern different linguistic structures, while the sixth heavily relies on extralinguistic knowledge. Nevertheless, it should be pointed out that the boundaries between these groups are not clear-cut, since different
4 The detailed overviews of services along with explanations and details on loan repayment, interest rates, additional benefits, terms, etc. do not constitute an advert anymore, and as such could be analysed separately.
MME1	LANGUAGE 33
levels of language are often simultaneously activated for a single pun to be understood, e.g. graphological, morphological, lexical, syntactic, and textual.
5.1 Graphological Structure
Puns based on the bank name incorporated with other words were found only in the Serbian corpus, as illustrated in examples (1) and (2). The name of the bank is followed by the service, merging into the slogan or advertising phrase:
(1)	OTPakujte vaše račune	[Unpack your accounts]
(2)	OTPlacujte stan sa lakocom!	[Repay your flat with ease!]
The wordplay is based on grapho-morphological merging. Unlike English, Serbian has a simple and logical spelling system, where each of 30 letters in the alphabet corresponds to one sound. In Example 1, OTP, the name of the bank, is combined with the word pakovati [to pack], to produce the new word otpakujte [to unpack; imperative, 2nd person plural]. In the second one, the compound consists of the bank's name and the word placati [to pay] resulting in otplacujte [to repay; imperative, 2nd person plural]. These new coinages indirectly imply that the bank easily enables those services. Generally speaking, coinages are created to give uniqueness, vigour and impact to the advertising message (Dyer 1982, 150). Using relevance theory and considering the fact that both advertisements promote bank services, the pun is used to provide additional positive cognitive effects and implicated contextual assumptions, which could be recovered as (i) and (ii):
i.	Our bank is always there to help you with important decisions (account activation, repayment).
ii.	These services are very simple and easy to get when compared to other banks.
In addition, the well-established acronym of the bank, PBB banka (Privredna banka Beograd), is cleverly incorporated in the slogan in the following two examples:
(3)	Planirajmo Bolju Buducnost	[Let's Plan Better Future]
(4)	Praktičnije Bolje Brže	[More Practical Better Faster]
The acronyms trigger two interpretations, both of which perfectly fit the context. The bank helps you plan a better future and is more practical, better and faster in terms of services and costs compared to others. These implications also create a positive image of the bank.
Another bank advertises a credit card called Panter [Panther]. The caption goes as follows:
(5)	Kupuj sigurrrno i nosipantera sa sobom! [Buy safely and carry a panther with you!]
The humorous reading is reinforced with the image of the person patting a real panther, as the card is both textually and visually presented as an animate entity.5 The caption further contains an onomatopoeic wordplay triggered by altered spelling, not reflected in the translation, since the word sigurrrno, features three r's, which imitate the sound of a panther angrily growling.
5 Along with puns, metaphors are very common in bank advertisements. However, they are not further addressed nor analysed in this paper.
34 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
Wordplay based on graphological structure is rather scarce in the British advertisements analysed in this study. There is, though, one that promotes a special credit card and a special current account with the same name — as shown in examples (6) and (7):
(6)	123 CREDIT CARD
Get cashback on everyday spending
(7)	123 CURRENT ACCOUNT
Switch your current account with our
5-Star Rated Switcher Service
The numbers are directly associated with the benefits obtained if you choose these services. Literally, it means that you will get 1—3% cashback depending on where you use the credit card (1% at all major supermarkets, 2% at all major department stores and 3% at all major petrol stations). Also, you can earn 1—3% cashback if you pay household bills using this current account. In these two examples the pun may not be immediately obvious, as the audience is satisfied with the first interpretation — the name of the service matches the amount of cashback. However, the effective combination of the benefits and the simile as easy as 123 enables another reading and expands the existing one. When this intended interpretation is recovered and processed, the following implicated contextual assumptions about the services are created:
iii.	It is extremely easy and simple to activate this current account / this credit card.
iv.	These services are both easy to get and use.
Yet again, the bank contributes to the creation of positive cognitive effects which reinforce a positive image of the advertised product and/or service. At the same time, the bank improves the chances of attracting more clients and promoting services in the best possible way.
5.2 Morphological Structure
As noted by Delabastita (1996, 130), in most morphological puns "words are construed as compounds or derivatives in a way which is etymologically 'incorrect' but semantically effective." No example was found to illustrate compounds which lost their original transparency, as in breakfast, for example. Instead, the puns in the corpora mostly rely on grapho-morphological structure and originate from the assumption that the given words are analysable units sharing the same root or affixes. The following advertising messages in (8) and (9) illustrate this type of pun:
(8) Sve(t) na dohvat ruke!
implying two possible meanings referring to the use of cards:
1.	Sve na dohvat ruke! [Everything at your fingertips!]
2.	Svet na dohvat ruke! [The world at your fingertips!]
Although sharing the same letters, sve [everything] and svet [the world] are unanalysable units, which are not related. Cleverly disguised into the message aiming to promote a credit card, this pun calls for two interpretations with clear positive cognitive effects being:
MME1	LANGUAGE 35
v.	With our card you can buy everything all over the world.
vi.	You can use the card anywhere.
vii.	Our bank offers you everything.
Furthermore, the unrelated free morphemes fun and fund are brought together in (9):
(9)	Putting the fun into fundraising
The advertiser emphasises this link while assigning new meanings. As a result, the audience may think of fundraising as an enjoyable and pleasant activity, and hence encouraged to collect and donate money.
In addition, new word combinations in bank offers are also produced by adding a letter to the existing word, as shown in examples (10) and (11):
(10)	Bez brige. Kredit je (o)siguran. implying two possible meanings:
1.	Bez brige. Kredit je siguran. [No worries. The loan is safe.]
2.	Bez brige. Kredit je osiguran. [No worries. The loan is insured.]
(11)	Proveri pa poveri
implying two possible meanings referring to the savings — first, check terms, offers and rates of other banks and then entrust your money:
1.	Proveri	[Check]
2.	Poveri	[Entrust]
Such grapho-morphological combinations additionally reinforce multiple readings and maintain multilayered communication between the advertiser and audience.
5.3 Lexical6 Structure - Polysemy
Myers (1994, 66) observes that wordplay based on polysemy is far more common in advertising than that based on homonymy. Puns based on different meanings of polysemic words are ubiquitous in the corpus. Moving from literal to figurative meaning back and forth, these words trigger multilayered communication and thus positive associations. The following example (12) in Serbian illustrates this:
(12)	Sve karte su u vasim rukama	[All cards are in your hands]
The noun karte [cards] can refer literally to debit / credit cards the bank offers and the caption would then read: you hold all the cards in your hands. This interpretation would produce positive cognitive effects, such as (viii) and (ix), and maybe even (x):
The term lexical is used here in a broad sense, including both the forms and meanings of words. In terms of punning, lexical items are both carefully chosen and deliberately manipulated to trigger multilayered meanings. Although I am fully aware that the terminology is not the most appropriate for the analysed semantic phenomenon, I decided to follow the classification found in Delabastita (1996). I would like to thank the reviewer for their insightful comments on this point.
36 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
viii.	I can use the cards quickly and easily.
ix.	The cards are always there when I need them.
x.	This bank offers everything I need.
The whole phrase can convey a figurative meaning 'to have control over your decisions', further implying something along the lines of (xi) and (xii):
xi.	You have full control over your finances if you use our cards.
xii.	Our bank helps you decide what is best for you.
The full potential of the pun is reached when both literal and figurative interpretations are simultaneously activated.
Moreover, there is an interesting advert for an energy efficiency loan, as shown in example (13):
(13)	Bezpromaje — i u stanu i u novcaniku
[No draught7 — neither in your flat nor in your wallet]
The first part of the sequence is interpreted in its literal sense, because a draught occurs when doors and windows are not properly sealed. Reducing air leakage in your home prevents excessive energy loss, but also reduces bills and minimises costs. The 'saving money' interpretation is activated with the second part, which is rather challenging for translation, since the idiomatic phrase promaja u novcaniku in Serbian means 'the wallet is empty', i.e. a person is penniless, but to the British a draught in your wallet does not evoke the same association and as such would be confusing. Therefore, the implication is that your wallet will not be empty and you will save money.
The same process takes place in British bank offers, which is illustrated and explained in examples (14) and (15):
(14)	New mortgage? Good move.
See if we can find you a mortgage that will save you money
In this example, the polysemous nature of the noun move enables several messages to be communicated. Let us start with the more obvious ones, as in (xiii) and (xiv):
xiii.	Taking out a new mortgage at our bank is a good decision.
xiv.	The process of changing homes will be smooth and will not cause any problems.
Both interpretations evoke positive associations and neither of them is rejected. Rather they complement each other since both are equally relevant for the audience. Furthermore, there is a possibility that this pun can also activate some deeper meanings, as suggested in (xv) and (xvi):
xv.	A new mortgage will improve your way of living.
xvi.	You will progress with a new mortgage.
7 The myth surrounding a draught is a socio-cultural phenomenon in Serbia since people tend to believe that it causes various problems (health problems included).
MME1	LANGUAGE 37
The phrase good move can also be connected with playing chess, hence it could be further speculated that the advertiser wanted to activate the concept of winning, since good moves increase the chances of winning both in sports and life.
Another example of polysemous words and phrases in the British offers is shown in (15):
(15)	Your account at your fingertips With our app or by text message
The caption includes the phrase at your fingertips, which can be understood literally — using your fingertips to type necessary information and click on the mobile app in order to access your account. When advertising mobile banking services, the intended meaning is more probably figurative — to have all the information about your account readily available and easily accessible.
Given the above-mentioned examples, it may be wrongly assumed that a pun almost always communicates two (or more) meanings. In the next example, one interpretation is rejected as irrelevant:
(16)	Sainsbury's Nectar Credit Card Sweet shopping
The pun in example (16) centres around the adjective sweet, which in combination with the noun shopping triggers the figurative meaning to describe the activity which makes you feel happy and satisfied. The literal meaning of sweet is triggered in relation to the card's name Nectar, only to be rejected as irrelevant.
5.4 Lexical Structure - Fractured Phraseological Units
Another very common technique found in the analysed corpora is a play on phraseological elements, i.e. the use of well-known phrases, where fixed word combinations are altered in order to generate certain meanings by different modifications. Beasley and Danesi (2002, 123) claim that the clever use of phraseology both imparts a sense of proverbial qualities to the product and generates a high connotative index. Myers (1994, 73) also mentions fractured collocations, where one word is deliberately replaced or a collocation is used in a new context, thus evoking specialised meanings and settings. Examples found in the corpus illustrate the modifications of fixed phraseological units, which are either:
a) associated with financial products, and hence given new interpretations:
(17)	Ima leka za dobro le enje. Finansijskapodrska za medicinske ustanove. [There is a cure for good treatment. Financial support for medical institutions]
(18)	1 za sve
Sakupite sve svoje dugove u 1 kredit ipla ajte manje! [1 for all
Collect all your debts into 1 loan and pay less!]
38 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
(19)	All's not lost (even when your card is)
(20)	Say no more to card usage fees
In the Serbian examples (17) and (18), the well-known idiom ima leka [there is a cure] refers to the bank's financial support, which can help medical institutions provide good treatment, while the part of the musketeers' motto one for all' is unexpectedly linked to one loan and all the debts which are to be paid back. Similarly, in the English example (19), the well-known saying all's not lost suggests that the situation can be improved, in this case when the card is lost. The informal8 phraseological unit say no more in (20) implies that all the hidden meanings are understood and no more details are needed. Referring to additional fees, the following interpretation can also be derived — you will no more have to pay card usage fees.
b) slightly reworded within the bank advertising message:
(21)	Sve vam jepopušteno % [Everything is discounted % for you]
(22)	Od duplogglava ne boli! [Double never hurts!]
(23)	Bank your way with HSBC
The past participle in example (21) within the phraseological units sve vam je oprošteno [you are forgiven everything] or sve vam je dopušteno [you are allowed everything] is changed to popušteno [discounted] and reinforced by the percentage sign, with the effect of attracting clients on the basis of a discount. In addition, example (22) in Serbian is a modified version of the idiom od viška glava ne boli [a little extra never hurts] applied to the interest rate for savings, which is doubled as part of the bank's special offer. In their advertisement shown in example (23), the British bank changes the collocation make your way into bank your way, suggesting that you will progress if you use their services.
5.5 Syntactic Structure
Syntactic ambiguity occurs when phrases or sentences can be parsed in more than one way, and it relies on the syntagmatic links between lexemes. For instance, the advertisement for a child's savings account reads as follows:
(24) Stednja za decu Blago deci! [Savings for children Treasure to children! / Lucky children!]
On a surface level, the purpose of the caption in example (24) is twofold: (a) to promote savings and (b) to encourage parents to open such an account. However, by analysing its syntactic structure further, new interpretations appear. The slogan blago deci can be analysed as a noun (nominative case) + noun (dative case) combination. The noun blago [treasure] can refer to money, gold, jewellery
8 Informal words and phrases are otherwise very rare in the register of bank advertising (Adams and Cruz Garcia 2007).
MME1	LANGUAGE 39
or any accumulated wealth, but also figuratively, to prosperity and well-being. On the other hand, this combination can be analysed as an interjection + noun (dative case), which is used to express satisfaction, happiness and joy. Although it can also be used ironically to show envy and jealousy, this interpretation is rejected here. With a recovery of syntactic relations, the pun provides access to contextual assumptions, which contribute to several positive cognitive effects, among others:
xvii.	By saving in our bank you will accumulate money for your children.
xviii.	Your children are lucky because you save for them.
xix.	Your children will have treasure.
xx.	Your children will be prosperous.
In example (25), the advertiser plays with the word bank. The message is also visually divided in two lines:
(25)	Bank with the best...
Best Current Account Provider
After reading the first line, the audience is very likely to process it as a noun + prepositional phrase, where dots create an anticipation effect. The idea is to make the reader fill in the gap with a service or a product the bank offers (e.g. Bank with the best interest rates / terms and conditions etc.). In other words, the reader wonders about the reasons why this bank is the best. Moving to the second line, there is a clear shift into a verb + prepositional phrase. The first interpretation is immediately rejected.
5.6 Shared Knowledge of Culture and Current Events
Finally, there is wordplay based on extralinguistic knowledge, where some influence of cultural elements and current affairs can be recognised. In Serbia, for example, politics is one of the most popular topics for discussion, so banks get their message across by reflecting and referring to political affairs, which are interwoven with everyday life, as in example (26):
(26)	Koalicija dobrih uslova	[A coalition of good terms]
This was the advertising message for a savings account during a post-election period, when the coalition agreement between two political parties was expected. Hence, the word koalicija [coalition] can refer both to the government formation of the two parties or literally, to the synthesis of good terms the bank offers.
In addition, the services are offered in connection with current events, as is the case with the beginning of the academic year for students in October. One advert for the bank account, specifically designed for students and their needs, contains the message:
(27)	Pravo je vreme za desetku	[It's the right time for ten]9
where desetka [ten] refers both to the highest student's mark and to the name of the student's account.
9 In Serbia, the grading system at the tertiary level depends on the number of points obtained during the course and at the exam, ranging from 5 (insufficient) to 10 (excellent — exceptional).
40 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
E L O P
In example (28), the advertisement draws upon popular cultural knowledge and stereotyping:
(28) How do you like your bank account?
Because everyone's different, we offer a broad range of current account to suit
every taste.
The bank offers different current accounts by asking potential clients How do you like your bank account? along with the image of a teapot. Another interpretation is triggered only when a verbal message is interpreted along with the visual. The visual component is crucial here, since it evokes a habit of drinking tea and suggests the alternative readings, such as How do you like your tea?, making a parallel with the kind of current account that would suit the client's taste. Therefore, a common British ritual is used to advertise the bank service. This causes the audience to use extra processing effort if they are to recover the alternative interpretation.
It should be noted, though, that the puns in (26)—(28) rely on shared knowledge between the advertiser and the audience in terms of cultural values, lifestyles and current events. Double interpretation is activated if and only if such knowledge exists. Otherwise, the pun would be unsuccessful and the intended message unclear to the audience.
6 Conclusion
The analysis presented in this paper revealed how the two languages are creatively used and how semantic rules are frequently broken in advertising. It was once again demonstrated that advertisements extensively use wordplay to maximise contextual implications and create a positive image of the focal product, regardless of language and culture. Tanaka (1994, 82) thinks that these language devices attract and retain attention, while Beasley and Danesi (2002, 102) claim that if the advert's textuality is ambiguous, it is more likely to create greater interest and generate appeal. As a consequence, this may affect the audiences' attitudes to the advert, and ultimately, the advertised product.
With regard to the research questions, this research study clearly demonstrated that puns are very useful attention-getting devices and that playing with linguistic and extralinguistic material is an effective advertising technique in both the British and Serbian bank offers. Both languages exploit the potential of punning even in more serious contexts, i.e. online bank advertisements, which is in contrast with the results for press advertisements for financial products reported by Adams and Cruz García (2007). Formal and semantic aspects of lexical units are creatively used on different levels. Interestingly enough, Serbian proved to be equally productive in terms of punning when compared to English, unlike some other languages (for a comparison between English and Spanish cf. Díaz-Pérez 2012). The findings also showed that the Serbian advertisements exploited graphological potential more than the British offers. The most frequent puns found in both corpora were complex ones with at least two interpretations being simultaneously activated. Sometimes, though, shared knowledge between the advertiser and the audience is crucial for deciphering and understanding a pun. In addition, some of the examples presented in this study show that puns do not necessarily fulfil a humorous function, but primarily capture and hold the attention of the audience, which is in line with the AIDA model.
This paper also demonstrated how relevance theory can be used to explain puns as ostensive stimuli in advertising discourse. Since they produce additional contextual effects, puns serve an important function in bank advertising and largely contribute to the overall and intended
MME1	LANGUAGE 41
interpretations and effectiveness of the message. These additional positive effects justify all the extra processing effort demanded from the audience to recover the intended message. As a result, puns are likely to be remembered, and so can further create a positive attitude towards the focal product or service.
Finally, the examples in this paper were discussed in terms of their verbal component. Nonetheless, since images are equally important message carriers, future research could address visual puns and examine how these can reinforce or change intended meanings triggered by the accompanying verbal text, and how they function as ostensive stimuli. The image used can narrow down the possible choices of word meanings, encourage particular readings and help solve double interpretations. Therefore, it would be of great interest to examine the changes in meaning prompted by the accompanying visuals.
References
Adams, Heather, and Laura Cruz García. 2007. "The Advertising of Financial Products in the Press." Ibérica 13: 123-46.
Albers-Miller, Nancy D., and Robert D. Straughan. 2000. "Financial Services Advertising in Eight Non-English Speaking Countries." International Journal of Bank Marketing 18 (7): 347-58. https://doi. org/10.1108/02652320010359561.
Beasley, Ron, and Marcel Danesi. 2002. Persuasive Signs: The Semiotics of Advertising. Berlin: Mouton de Gruyter.
Byrne, Barbara. 1992. Relevance Theory and the Language of Advertising. Dublin: Trinity College Dublin.
Caples, John. 1974. Tested Advertising Methods. Fourth Edition. Englewood Cliffs, NJ: Prentice-Hall.
Choi, Jieun, Charles R. Taylor, and Doo-Hee Lee. 2017. "Do Resonant Advertisements Resonate with Consumers?" Journal of Advertising Research 57 (1): 82-93. https://doi.org/10.2501/JAR-2017-007.
Collins Cobuild Online Dictionary, https://www.collinsdictionary.com/
Cook, Guy. 1992. The Discourse of Advertising. London: Routledge.
Crystal, David. 2003. The Cambridge Encyclopedia of the English Language. Second Edition. Cambridge: Cambridge University Press.
Delabastita, Dirk. 1996. "Introduction." The Translator 2 (2): 127-39. https://doi.org/10.1080/13556509.19 96.10798970.
Díaz-Pérez, Francisco Javier. 2000. "Sperber and Wilson's Relevance Theory and its Applicability to Advertising Discourse. Evidence from British Press Advertisements". Atlantis 22 (2): 37-50.
—. 2012. "The Use of Wordplay in Advertisements Published in Men's Magazines. A Comparative Study in the UK and Spain." Estudios Ingleses de la Universidad Complutense 20: 11-36. https://doi.org/10.5209/ rev_EIUC.2012.v20.39993.
Djafarova, Elmira. 2008. "Why do Advertisers Use Puns? A Linguistic Perspective." Journal of Advertising Research 48 (2): 267-75. https://doi.org/10.2501/S0021849908080306.
Dyer, Gillian. 1982. Advertising as Communication. London: Routledge.
Grice, H. Paul. 1989. Studies in the Way ofWords. Cambridge, Massachusetts: Harvard University Press.
Horky, Vit. 2009. "Guerrilla Marketing Online." Accessed April 22, 2013. http://www.guerrillaonline.com/cs/ Guerrilla-Marketing-Principles-54.htm
42 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
Jewler, A. Jerome. 1992. Creative Strategy in Advertising. Fourth Edition. Belmont, California: Wadswort.
Koren, Ana. 2012. "An Honest Attempt to Grasp and Possibly Tame the Wild Animal of Punning Taxonomy." ELOPE 9 (2): 23-37. https://doi.org/10.4312/elope.9.2.23-37.
Kotler, Philip, and Gary Armstrong. 2012. Principles of Marketing. 14th Edition. Boston: Pearson Prentice Hall.
Labrador, Belen, Noelia Ramón, Hector Alaiz-Moretón, and Hugo Sanjurjo-González. 2014. "Rhetorical
Structure and Persuasive Language in the Subgenre of Online Advertisements." English for Specific Purposes 34: 38-47. https://doi.org/10.1016/jxsp.2013.10.002.
Laviosa, Sara. 2005. "Wordplay in Advertising: Form, Meaning and Function." Scripta Manent 1 (1): 25-34.
Lazovic, Vesna. 2014. "The Language of Online Bank Advertisements in English." ESP Today 2 (1): 88-104.
Leech, Geoffrey. 1966. English in Advertising: a Linguistic Study of Advertising in Great Britain. London: Longman.
Leigh, James H. 1994. "The Use of Figures of Speech in Print Ad Headlines." Journal of Advertising 23 (2): 17-33.
McGuire, William J. 2000. "Standing on the Shoulders of Ancients: Consumer Research, Persuasion, and Figurative Language." Journal of Consumer Research 27 (1): 109-14. https://doi.org/10.1086/314312.
McQuarrie, Edward F., and David G. Mick. 1996. "Figures of Rhetoric in Advertising Language." Journal of Consumer Research 22 (4): 424-38. https://doi.org/10.1086/209459.
Myers, Greg. 1994. Words in Ads. London: Arnold.
—. 1999. Ad Worlds: Brands, Media, Audiences. London: Arnold.
Ogilvy, David. 1964. Confessions of an Advertising Man. New York: Atheneum.
Partington, Alan S. 2009. "A Linguistic Account of Wordplay: The Lexical Grammar of Punning." Journal of Pragmatics 41 (9): 1794-809. https://doi.org/10.1016/j.pragma.2008.09.025.
Redfern, Walter. 1984. Puns. Oxford: Blackwell.
Sherzer, Joel. 1985. "Puns and Jokes." In Handbook of Discourse Analysis, vol. 3: Discourse and Dialogue, edited by Teun A. van Dijk, 213-21. London: Academic Press.
Sperber, Dan, and Deidre Wilson. 1987. "Authors' Response." Behavioral and Brain Sciences 10 (4): 736—54.
—. 1995. Relevance: Communication and Cognition. Second Edition. Oxford: Blackwell Publishers.
Sternkopf, Sylva-Michele. 2005. English in Marketing: International Communication Strategies in Small and Medium-Sized Enterprises. Berlin: Frank & Timme.
Strong, Edward K. 1925. The Psychology of Selling and Advertising. New York: McGraw-Hill.
Tanaka, Keiko. 1992. "The Pun in Advertising: A Pragmatic Approach." Lingua 87 (1-2): 91-102. https:// doi.org/10.1016/0024-3841(92)90027-G.
—. 1994. Advertising Language: A Pragmatic Approach to Advertisements in Britain and Japan. London: Routledge.
Thaler, Verena. 2016. "Varieties of Wordplay." In Crossing Languages to Play with Words. Multidisciplinary Perspectives, edited by Sebastian Knospe, Alexander Onysko and Maik Goth, 47-62. Berlin, Boston: De Gruyter. https://www.degruyter.com/view/product/468240
Tom, Gail, and Annmarie Eves. 1999. "The Use of Rhetorical Devices in Advertising." Journal of Advertising Research 39 (4): 39-43.
van Mulken, Margot, Renske van Enschot-van Dijk, and Hans Hoeken. 2005. "Puns, Relevance and
MME1	LANGUAGE 43
Appreciation in Advertisements." Journal of Pragmatics 37 (5): 707-21. https://doi.org/10.1016/j. pragma.2004.09.008.
Vestergaard, Torben, and Kim Schroder. 1985. The Language of Advertising. Oxford: Basil Blackwell.
Wilson, Deidre, and Dan Sperber. 2004. "Relevance Theory." In The Handbook of Pragmatics, edited by Laurence R. Horn and Gregory Ward, 607-32. Oxford: Blackwell.
44 Vesna Lazovic How to P(l)ay with Words? The Use of Puns in Online Bank Advertisements in English and...
Jakob Lenardič
University of Ljubljana, Slovenia
2018, Vol. 15 (2), 45-61(120) E
revije.ff.uni-lj.si/elope doi: 10.4312/elope.15.2.45-61 UDC: 811.111'367:811.163.6
The Middle Construction in Minimalist Syntax: English and Slovenian
ABSTRACT
The paper discusses the argument structure of the English middle construction and its Slovenian equivalent from the perspective of minimalist syntax. The paper first introduces Bruening's (2012) recent approach to syntactic middle formation, which posits that middle sentences are derived via an operator that existentially quantifies over the open agent variable introduced by an active Voice projection. Subsequently, the paper argues that the adverbial modifier in the middle construction is not a semantic argument of the null operator, contra Bruening (2012). Finally, the paper proposes that the reflexive morpheme se in the related Slovenian se-sentences plays a role of valency reduction similar to that of the null English operator.
Keywords: valency; middle construction; reflexive morpheme; syntactic minimalism; argument structure
Srednjiška zgradba v minimalistični skladnji: angleščina in slovenščina
POVZETEK
Pričujoči prispevek obravnava vezljivost angleške in slovenske srednjiške zgradbe z vidika minimalistične skladnje. Prispevek najprej predstavi Brueningovo (2012) skladenjsko razčlembo srednjiške zgradbe, kjer ključni del skladenjske izpeljave sloni na fonološko ničtem skladenjskopomenskem operatorju. V formalnopomenskem smislu je tovrstni operator funkcija, ki uvaja eksistencialni kvantifikator, ta pa se v pomenski ravnini veže s spremenljivko, ki jo uvaja jedro zveze glagolskega načina Voice. Prispevek tudi predloži, v nasprotju z Brueningom (2012), da prislovna zveza v srednjiški zgradbi ni pomenski argument ničtega operatorja. V povezavi s slovenščino prispevek argumentira, da prosti morfem se igra podobno vlogo v povezavi z vezljivostjo kot angleški ničti operator.
Ključne besede: vezljivost; srednjiška zgradba; povratni morfem; minimalistična skladnja; argumentna struktura
MME1
LANGUAGE 45
The Middle Construction in Minimalist Syntax: English and Slovenian
1 Introduction
The argument structure of the middle construction (1) has been a long-standing grammatical puzzle.
(1)	Politicians bribe easily.
The main point of contention in the formal literature relates to the syntactic representation of the implied agent argument of (1); that is, the person or people who can easily bribe the politicians. The problem is that the construction behaves inconsistently with respect to tests that diagnose the structural presence of the agent. On the one hand, the middle sentence in (1) differs from a prototypical passive construction (2) in that it cannot licence the by-phrase (3), by means of which the implied agent can be represented overtly.
(2)	The politician was bribed (by John).
(3)	Politicians bribe easily (*by John).
What is puzzling, however, is that middles are able to select an instrumental with-phrase (4), which is also allowed in the passive (5), but not in the ergative anticausative construction (6), which is standardly analysed as completely omitting the agent argument in its syntactic, as well as semantic, representation (for an overview of recent approaches, see Alexiadou Anagnostopoulou, and Schäfer 2015, Chapter 3). In this sense, it is often observed that the agent in the middle is only "implicit" (Klingvall 2005), so absent from the syntactic derivation.
(4)	The door opens easily with a key.
(5)	The door was opened with a key.
(6)	The door opened (*with a key).
The cross-linguistic grammatical properties of middles differ widely from language to language (Lekakou 2004). For instance, the Slovenian middle equivalent of sentence (4), which is given in (7), contains in its structure an additional category in the form of the reflexive morpheme se.
(7)	Vrata se zlahka odprejo s ključem. doors.NoM refl easily open with key "The door opens easily with a key."
Apart from se, Slovenian middles differ from English ones in that they do not involve the so-called generic repair strategy (Härtl 2012), which means that the majority of predicates are licit in the English middle construction if and only if the interpretation is generic (8b), whereas the equivalent of eventive (8a) is perfectly fine in Slovenian (9a).
(8)	a. *The bread cut.
Intended: "Somebody cut the bread." b. The bread cuts easily.
(Levin and Rappaport-Hovav 1995, 95)
46 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
(9)	a. Kruh	se je rezal.
bread.NoM refl aux cut
"Somebody was cutting the bread."
b. Kruh	se zlahka reže.
bread.NOM refl easily cut
"The bread cuts easily."
The aim of this paper is to discuss how middle sentences can be syntactically accounted for, taking into consideration both their selection properties stated in (2—6) with respect to the implicit agent and the cross-linguistic structural properties that distinguish English middles from Slovenian ones (7—9). In terms of theory, the paper will follow a syntactic approach in the spirit of constructional minimalism (Ramchand 2008, Bruening 2012, Alexiadou, Anagnostopoulou, and Schäfer 2015, to name a few), in which the formation of lexically complex predicates, such as those in the middle construction, takes place entirely during the syntactic derivation. Within this framework, the paper will first present and discuss the empirical and theoretical advantages of Bruening's (2012) recent proposal of syntactic middle formation in English, where the construction is derived by means of a null operator that targets the open agent variable that is introduced by a Voice head. Subsequently, the paper proposes that the adverb, although obligatory in many middles, is not a semantic requirement of the null operator, contra Bruening (2012). Finally, the paper will discuss how Slovenian se-sentences are derived, and propose that se plays a similar role in semantics to the English middle operator, but has fewer selectional requirements. In this respect, the paper will also propose that sentences (9a) and (9b) are structurally the same (i.e., se has the same function in both sentences, in that it binds the open agent variable introduced by Voice), with the only difference being that the event variable in (9b) is bound by a generic quantifier.
The paper is organized as follows. Section 2 lays out the theoretical framework and presents Bruening's (2012) proposal as an alternative to the traditional lexical account of middle formation, and discusses how his syntactic approach elegantly accounts for the key properties of English middles, such as the arbitrary reference of the agent, the generic interpretation, and the licensing of the with-phrase. Section 3 discusses the role of the adverb in the middle construction in relation to Bruening's (2012) operator. Section 4 discusses related Slovenian se-sentences and proposes a semantics for se contrasted with the English null middle operator. Section 5 is the conclusion.
2 Deriving Middles - From the Lexical Approach to a Minimalist Syntactic Account
2.1 Middle Formation in a Lexical Framework
The traditional approach to the derivation of the middle construction is the lexical account, in which the middle construction (11) is understood as a pre-syntactic derivational transformation (i.e., detransitivisation) of an otherwise active transitive sentence (10).
(10)	John reads this book easily.
(11)	This book reads easily.
MME1	LANGUAGE 47
Ackema and Schoorlemmer (1994, 65—67) describe a very detailed implementation of such an idea, which — at least to some degree — informs much subsequent work on middles (e.g., Marelj 2004 for Slavic). The gist of their proposal is that there is a pre-syntactic level of representation called Lexical Conceptual Structure, which contains lexical rules that can modify the argument structure of a predicate like read before syntactic projection takes place. In this sense, they propose that there is a rule called Middle Formation that pre-syntactically transforms an otherwise active transitive entry, (12), into a middle one, (13).
(12)	readactive (OActor' ^patient^
(13)	readmiddle (A.RB, ^J)-
Such a lexical approach is conceptually very appealing, as it accounts for a host of disparate grammatical facts of the middle construction. For instance, the transformation of the lexical entry read from (12) to (13) shows that the Middle Formation is an operation in which the agent 0-role is pre-syntactically realised as a non-projecting arbitrary participant (ARB in (13)). Since the agent is thereby saturated in the lexicon, it is not present in the syntactic component, which explains why middles cannot license the by-phrase, whose admissibility is otherwise assumed by the lexical approach to be contingent on some kind of structural presence of the external argument (e.g., for a passive sentence, which admits the by-phrase in contrast with a middle, Marelj (2004) assumes that the external argument is present in the syntax as a demoted oblique argument).
Notice, lastly, that lexical middle formation posits an argument-linking rule called Externalize, abbreviated with the superscript Ext in (12) and (13). This rule entails that the remaining 0-role is realised VP-externally; that is, as an external argument. The rule is assumed in order to explain a host of cross-linguistic data, such as the fact that middles morphosyntactically pattern with the intransitive unergative construction, which selects an external argument. For instance, in Dutch, middles — like unergatives — select the auxiliary verb have (14) in periphrastic tense constructions, whereas unaccusatives, which realise their sole argument vP-internally, select be (15).
(14)	Dit vlees heeft/*is altijd gemakkelijk gesneden. (Dutch middle) this meat has / is always easily	cut.
"This meat has always cut well."
(15)	De spanning is/*heeft dagenland toegenomen de spanning. (Dutch unaccusative) the tension is/ has for-days increased.
"The tension has increased for days."
(Examples taken from Ackema and Schoorlemmer 1994, 61, 64)
2.2 Introducing the External Argument in a Minimalist Theory of Grammar
In contemporary theories of grammar, however, there has been a longstanding aim to reduce the complexity of such a model — i.e., to make the grammar "more minimalist" in Chomsky's terms (2001) — by placing all the compositional burden of word-formational processes on the syntactic component (e.g., Marantz 1997; Ramchand 2008). One of the problems of a non-minimalist grammar that employs pre-syntactic rules for altering argument structure is that it invariably
48 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
leads to what Ramchand (2008, 15) calls "massively stipulated homonymies"; in the case of middles, this is the idea that there are actually two separate entries for the verb read — one regular active (12) and one middle (13) — in the English lexicon, which is additionally suspect from the point of view of learnability (Borer 2004).1 Consequently, what is needed from the point of viewofsuch constructional minimalism isa theory ofmiddles in which the structuraland selectional quirks of the construction (i.e., the licensing of with-phrases and the inadmissibility of fy-phtusyu, their generic iateruyetafion, their patterning with unergatives) are all handled during the syntacticderivation.
I staei tlie disesssiod by logksng cl how tgs argument serucfuee tit e predlcaie is built in a minimalist thsow of language- Sinct she main cbncern is sUe stciun ob tha bpfene isi middles, the ensmciisi ssoisssiyiieiet dsns dtils g)ui]Cosd it Kiatsets (df96i^ inyutii^tiy proporal fisas lhe eilttsnal themustie rok oe a prcdivni^ nniei-s lhc sgn(aciic derivation cutaici(i the kpicel yreub Pheasea wtrich hosis soklythe iytemai;drgument.sn herpsoposeka verbhkt sedc(kxicaiiy (tauotesarelation betwoen an cvent oXreading and ae entity (i.e.,rhe patient) that is read in the event (16).
(t6) frepy] ^Ase(e(.Sl^a(s)i[rt^D(e) sn^ai^ient(x, al]
Note shai t(ce wcy tire iDntactic itrucrulr ts buiit classli>siletSeprt isf^ciemsntic titcjisii'ssae^ntii of a gpiven dradioatP.Conareteiy nhe meresr ot i)lií] iexicesl item endst esssrC ipse; iltP etc ¿>sc)f, whlch rerukt ic sle syntroticitmcture Iii (it7), sem;ultiasely corresdondstn OunctiDnaCapplicatiDn) In that: )iee verb re^st demdaescfunction (1_6) that selects the denotation of the DP the book as its first semantrc su^e^ni^ia^en^, as shown in (18).
(17) TPu san(ur aU Vt
(17) The semantics ofmergingV and DP
e [VP] = |OMDP]) = IeacUKI(Pa Soak]) = 0: Or^Oe^tpVADD) n PATiENTOr el])(Pu So ttk) = a. Oe^READOei n patient)(Pu S)oak( e)]
At thenextstep of structure building,the externalparticipant is introduced into thesemantics of VP. To this end, VPmergesinKratzer's system(1996)withafunctional head labelled Voice, asin (19). Semantically, this head expands the denotation ofVP, given in (18c), by transforming it into a two-argument functionthat selects anadditionalparticipantatthelevelofthe intermediate projection Voice'; since read lexicallydenotes adynamic event, the Voice head identifies this participantasthe agentofthe event.2
Such stipulated homonymy is even more apparent with causative predicates, where it is assumed in the lexicalist framework that there are three separate entries for the verb open — (i) the anticausative entry, as in This door openedandcausadve, created by the process of anticausativisation, which involves pre-syntactic external argument deletion (Levin and Rappaport Hovav 1995);
(ii) the middle entry, as in This door opensmiddle easily, created by Middle Formation as described in Section 2.1; and (iii) the default causative active entry, as in John opened ausative the door.
Some of the examples in the paper have bare plural nouns as internal arguments (e.g., Physics books read easily). Strictly speaking, such DPs are semantically complex functions, but for the purposes of this paper, I assume a simplified representation
MME1	LANGUAGE 49
(19) The syntax and semantics of VoiceP without an external argument VoiceP
[Spec VoiceP]
Voice' A.x.A.e.[AGENT(x, e) A READ(e) A PAITENT(the book, e)]
Voice
Xx.A£.[agent(x, e)]
VP Xe.[READ(e) A PATIENT(the book, e)]
V
read
DP
the book
To derive a prototypical transitive sentence such as John read the book, what the syntactic computation needsto do is simply merge the DP John\nthe [Spec,VoiceP] position to saturate thnopenvariable ofVoice'(20b). Thefinal somaneics for VoincP in g;iveo in (20c).
(20) The semaotips of VpiceP eskls exterpill argume^ John
a.	Sy2ipsC]= |C))oice ' ]] ( [[.Joleti]] ) =
b.	Xret.Xe^.lAGENhpx, s) a read(v) a CAhieNhhths book, s)](Joev) = p. Xe<s>. [AGENh(Johv, s) a read(v) d CAhiENh(ths book, s)]
Thereisan agreement inthe syntactic literature(e.g., Bruening2012,Legate20l4, Alexiadou, Anagnostopac^ulo^, and Sahäfer 20a5) ti^at Vp-externalvalency reduction (that is,theomission of ah overt exiernal ergemena m middlesianticousatives, pussives, ete.i oreurs at sXn eoege in whi^lbVoicoP is °uilt.The propcoa1 )hel I willnsw adaptfor m^dka Staken esona Bornening
Bruening suggeststhat the derivationof a middlesentenceproceedsinitially inthesame manner as that ofa prototypicalactivetransitive structure, in th_e sense that; a VP containingtheinternal argumunt iirte merget wiih a Voke heani introducing the open ageat neriable. Tfc juoioc of syntactic uspviliata in the middle centsnce, nowevef hrpjtens brcause ea mergeo a semantic opeoaiorthat aekcts an untaturated Vricepeu jection (i.e;, she toeiin^^tion(^f she mteameiate category Voice'in(19)),andtherebyblocksthemerger ofareferring DP as the external argument. Thecemantiarfoesuehes ctperaloo etenlvehie(20t.
Accordingto (21),MiddleOp selects asitsfirstsemanticargumenta syntactic categorythat logically corresponds to a function with an open agent variable, and maps it onto a scope where the agent variable is bound by an existential quantifier. The second argument fed into (21) is an eventmodifieroftype (s,t),whichcorrespondstotheadverbialmodificationinmiddles(i.e., an adverb like easily isinterpretedasafunctionfromeasy eventstotruth values;see the syntactic treein(22)below).
where theycombineviaFunctionalApplicationwith thepredicate inthesamemanner asthe simple definitedescriptor, whichisoftype (e),in(18c).
2.3 ASyntacticAccount of EngUsh Middlsb
(201S, 2eualli.
(2 S) SMiddlsOp] = 5tfi(s,(e,tii.e,tp.GNs3 x[f(x, s) -m S(n)]
(Bruening2012,31)
50 Jakel IisbatdiciVilVsn/¡CeœiPva2s« in MtsimatistPnntas:EngWe ordShvensan
Syntactically, note that Bruening's (2012) proposal is cast within a theory of bare-phrase structure where operators can be merged before a grammatical head — Voice, in this case — fully projects. Differently put, the semantics in (21) suggests that MiddleOp is merged in the [Spec, VoiceP] position, but this position is canonically (e.g., in X-bar theory, whose notation the paper has partiidly adopted) associatedwithnominalexpressions(DPs saturating the agent variable), and noewith operatorrthat (electehein own semantic arguments. aMtheugh such ae assumption is emgiriaally and thenresically seeii mowioatea(see thr Uiseussion in Bautaicg 2W12, tO^eueS, it seense liesO itiid whole derivaiioa can U)e simphnrn — a( Seds( dsn eese oe axposilton -s ty esnureing thai MlddteOp hradt its ouen jrlitatd EtitLitl seSertn s VdSaeP wPich hes eesto meagaU an sxteenal argewmetit its iau suraifiet stot.Ttat is, the VP ofa midaieeonnsrucironmerges with £t apecial Voice aaad,wticO intaoduces an open agent variable, but does not require a filled specifier slot.3 I thus propose that an English middle construction has the syntactic representation in (22).
(22)	MiddleOpP
MiddleOp	VoiceP Xx.Xe.[AGENT(x, e) A READ(e) A PATIENT(the book, e)]
ApplningMiddleOpto che specifier-less Voice projectton, the syntacticdetlvation arrives at the semantic representrtionof tpe ottddte sentence TOlshook rradseasilyeioax is given inÇ03).
(23) Theinterpretation ofThisbook readseasily
a. [MirdtrPo] ( HVcitrP] )( [parity ] ) =
It. fteed>.Xpott>.GNrtx[f(x, e) —t =(e)](^x<r>.^e<r>.[agent(Xt r) a read(r) a p'/atient(thr Ibcckt p)]) (ta<r>.[easy(p)]) =
t. GNr3x[[agent(Xt r) a read(e) a patlent(ihr bcckt r)] — [easy(r)]]
Since the existentially bound agent variable is within the scope of a generic quantifier, i.e. GN, thefinal interpretationofa sentencesuch as This book readseasily,givenin(23c), is Generally, wheneversomeonereads the book,shedoes soeasily.
Such a semantics and its corresponding syntactic derivation account for many of the properties of middles discussed in Section 2.1. First, recall that, cross-linguistically, middles do not patternmorphosyntacticallywithunaccusativesentences,which mergetheir sole argument VP-internally, butratherwithunergative sentences,whichmerge theirargument VP-externally, in thattheyselect theauxiliary have insteadof be inperiphrastic tenseconstructionsinlanguages such as Dutch (examples (14)—(15)). On the current account, the patterning with unergatives simply followsfromthe fact thatmiddles containa Voice projection whosehead semantically brings abouttheexternal participant.Such a VoiceP isalso presentinunergativesand transitives,
3 This assumption is in line with recent work on the grammatical properties of the Voice domain (e.g., Legate 2014; Alexiadou, Anagnostopoulou,and Schäfer 2015),whosecentralclaim is that languages employ a set of different Voice heads with unique selectional(e.g., a Voicehead mayormaynotintroduce the external participant variable) and syntactic (e.g., a Voice projection may or may not require a filled specifier slot) properties.
E
L
□
P E
MME1	LANGUAGE 51
but not in unaccusatives, which is why they select be. Furthermore, note that the English middle construction can only be headed by a lexically transitive predicate (24a), in contrast with a prototypical unaccusative sentence (24b).
(24)	a. Obedient daughters raise/*rise easily. b. The sun rises/*raises in the east.
(Examples from Schäfer 2007)
An additional advantage of this syntactic approach is that it does notneed topositaspecial lexical operation whereby the external participant in the middle is realised as an arbitrary agent (Fagan 1992; Ackema andSchoorlemmer 1994; hekahau 2004;Marelj 2004). Thearbitrary flavour simply comes from the fact that the existentially bound agent variable x occurs within the acoue of the geneoic 2peratoc(which binds a sicuationnoriabae);ihat Is, thefinal LFof a mihdlosentencein(23r),whirh2orrespondsto tCio parayirate Generally, wheneeeraomeooe heads thn bttk, s/SaLSaf?t.te^•f«7e, ii ^n^^rp^^eeti^i^o^eoke^)'i^mii^it^oiSag^n's(l^i)S^^2) ]^j^(^]posedj:^a^;^j^lbrase Anyone cap read2hisboo( rasiPy,'mt\it senserhat "anyone" isnimply"someone"who deeaersfrom onigiinridituationtoanother. Conceptually, rhln,sucn a syntaetec sccount allows ds tederive a miSdtecenocnen withadtdrcoursc iea eran)format)ven4asl ln thelek^i^i^n tk^^r hee-syntoctlaally tucnsr vesO likerp•nal. whtch islnherenllylransesive,into a mid01evaripnt wstO aia iabebenyantly posiledas-bltralyagent cok (c; • ^-s1 epth
The greatest advantage of the derivational syntactic account, however, is that it offers a principled fo rmal exbtanetfon esto whya mcidle eentnnte licontet ce inttaumentai^ct¿sp-bhcks e. Re call th it, rn thiei^^!^f)iat, thementkseMs ncetn (2aLnpntSerns wiA sheragulalactivr tehtenne in (2bne nnt wsth tbc anoicautntive samcrnre iic ^^^ite^) formen from C2esamcpreacaaeo.
(25)	The licensing of an instrumental with-phrase
a.	Middle: This Oooe opens easily with a -ey.
b.	Active: John opened this Oooe with a -ey.
n. Anticausative: The Oooe opened (*with a -ey).
Brucrnng(20l2, 27)proposes that witi is inselpselattnely a Cune)ionihal selectlt0ndeho)ation of as syndetic nomptement DP, which demobes some iostsnment noebbntna agens as iesfirst semantic cnumnnt.
(2f7 iwith] = Xx<e>lf<e,<s,i>>.Xy<e>.Xe<s>.[f(y, e) a 3e'<e[üSlNG(x, s') a AGENT(o, s')]]
Let usassumethatthereisan event e in whichsomeoneopensthedoor.Thesemanticsin (26) say thatthereexistsasubevent of e — namely e —,in whicha personinterpretedasan agent y opens the door by using an instrument x. In other words, what the preposition with does semantically is it expands the denotation of the syntactic structure by introducing an additional participant that interpretatively corresponds to the instrument.
Syntactically, the keyideabehind (26) is thata with-phrasecanonlyattach to a structurewhich contains anopen agent variable, as seen in the factthat with obligatorily selectsas its second semanticargumentafunctionoftype (e,(s,t)). Onthesyntactic account,middlesareindeedbuilt ontopof sucha structure(VoicePin(22)),hence the reasonwhya with-phraseisadmissible. By contrast,ananticausativestructuresuchas (25c)isstandardlyanalysedaseitheromitting a Voice
52 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
projection or containing one that is semantically vacuous, which means that the denotation of VP is not expanded to contain an agent variable.4 The denotation of a non-expanded VP is of type (s,t) (a function from eventualities to truth values), as in (18c), so an instrumental PP cannot be structurally adjoined to it, as it would result in a type mismatch.
I believe that this is,from thepointofviewofminimalist theory, the bestway inwhich the InstrumentGeneralization, which states that "art [insfremen th PP] raquiresthe explicit(syntactic) or implecie(temqetic) preeence of en ngent in oeder ^cs be realizsd tpenrecticall)^"(c/[are[j 2004, 11(5), can be accounted for in or fotmal mannet. Hee ken adva^a; oe7 tire syntactic approach is thsf st re-sratrs thit salhee mformee — anrquite eegue — diitinction lietcvhtf an agencthat is pieaent on^)p un^icidy md iint;t^ltat ispreseci explmtly in muchcle avartCTms: theimjelicit agrntrna medrie as onethai hegin^ its g;t£rt:ran;^sic;ai life ye ap pijesn veriabk denotitte a veHdonal paaiicijiaaiti only tobe kid os eaturated by ecisteariai closura beoughs at>ant fn aa	euch
as (hfi. Hence,beefnsesheahrua innt one ¡^olntanopeat"^iteiab^e, tilt e^fa^^-jjtira^e^ ^dv^^e^d.
3	The Role of the Adverbial Modifier
In this section, I proposeirnamendment to Bruening's (2012) account related to the adverbial moelifire, which ia nanited tee a nacerstry requirement sf the rmddfe aonttaycriov bslause MiOn(oOp inteoduces the auntaia ofieeatoo	i^he Uuui'atu^^ant^^r^^i^ieiie^^l^ifka
et al. 1 999), n generis q^Mie^t GN ^t(t(iti(e:rit tire di od a j^^n^ria; oent^nee into two pyets — thr r^srracte)'^ anil elor; nuaiean eaeps. "t ls c neceetaep condidon ah ^iittt^ c tveo-niart LF dy the nueleae s(iotle hatst tempntic tis^ntent: t!Kolfle^ ^t th e99ti, )t">ti^/i)t wHdi tixjsiltaln^ why c senience likes Wis book readh dfeasify) is ill-fdemee1 lf rntin ir tmkted.
Genially sha temenitiao o,, iihi(^d(si3ji evoorualy prediice elsaia tha eJor'titioning *s teken case of by shetynttctia derivhskrn kailS;	by ahe fact enaatheoorresoe ^qatarae AaAdlP ep its
seconnsemanticargomcnt,whichet thenmeptini:o riac nedem sconoiI wiil now show thotauch a requeeement rs loo ererat tn accdnnt fer el EngitrO msddier: ac^ct tre^te its aormfe reperssnscaiun of MihderOushouiii fie rimjiUfiedfrum ks oe^inaHosme lation in(2].), repsaaed herein^/^so^8).
(2nf [MillleO)] td^nrual haolrdu- = Xf<a,<s,t».Xg<s,t>.GNa^x[f(x, e) ^ g(e)]
(28f IM"!!^)] (savlsaa vasslan) = f ^¿».GNiiaxnO^ e)]
Whatiskeytothesimplified semanticsin(28)isthattheadverbialmodifier,althoughobligatory in manymiddles,isnolongerasemanticargumentoftheoperatoritself.Therearetwopieces of evidence forthisassumptionandthe consequentrevision.First,therearemiddlesentences in English that are perfectly licit without an AdvP, as shown by the following two examples, taken fromMarelj(2004,125).
(29)	Thre vfeoalla folde up.
(30)	TqlsOsass buttans.
4	According toSchaferandVivanco (2016),ananticausative sentence like Thisdooropened simplymeans Therewasan event in
whichthis doorbecame open;since Voice iseithernotmergedorsemantically inert, nospecifictype ofexternalparticipation is
entailed. Thesyntacticandsemanticlackof anexternal participant makes the sentence compatible with describing a number ofpossiblesituationsinwhichthedoorbecomes open;the external participantmightbe anon-humancauser (e.g., Thewind openedthedoor), orthereeven mightnotbeanyapparentexternalcause,as isevidentbythefactthatananticausativesentence
admitsthe by-itselfphrase: Thedooropenedbyitself.
kANddAGE 53
Despite the absence of the AdvP, the LF of such sentences is still partitioned into a restrictive and nuclear scope, as required by the generic quantifier. According to Marelj (2004, 127—128), the LF partitioning has an extra-semantic source in contrast to the original semantics of MiddleOp in (27) — i.e., it is licensed by prosodic structure, in that "the focused part of an utterance with an operator such as [GN] is always in the [nuclear scope]" (Krifka et al. 1995, 27). What then licenses an adverb-less middle sentence like example (30) is the placement of focus stress on the verb buttonsQrAcConnell-Ginet 1994).S uchfo cus placement bringsabout the LFin ( 31), where (a part of) the denotation of the verb buttons is mapped into the nuclear scope.
(31)	GNe3x[[C(e) A agent (x, e) A theme (this dress, e)] ^ [buttoning(e)]
(AdaptedfromMarelj 2004, 127)
TheLF in(31), inwhich C is takentobea variablerangingovercontextually relevant information, canbe paraphrased as Wheneversomeone doessomething to this dress ina contextuallydetermined event, she buttonsit, where the contextually determinedevent is "mostlikely associated with someaspect ofxputting on a dress in anappropriateway" (Marelj 2004, 128).5 Such an LF, however, is not predicted by the original formulation of MiddleOp in (27), which banks on theassumption that theLFpartitionnecessarilyrequires thesyntactic presenceof an additional constituent (i.e., the adverbial modifier), which is then mapped into the nuclear scope. In other words, this assumption goes against the adverb-less middle example (30), as the partitioning of itsLFtakesplace withoutthe additional piece of structure.
The second piece of evidence lies in the fact that the operator's requirement for a second semanticargumentderivesword order thatistoorigid. Toseethis,recallfromSection 2.3 that an instrumental with-phraseenters thederivationas a Voiceadjunct. What is crucial is that this takesplace beforeMiddleOpenters the structure,giventhat with semantically needs an open agentvariable(26). Consequently,ifMiddleOphadthesemantics in(27),thewordorder would be invariably as in (32); namely, the AdvP would necessarily be higher in the structure than the with-phrasebecauseofthesecond semantic requirement of the operator, making the standard wordorder in (33) difficult, ifnotimpossible, to derive.
(32)	?Th)sdoosopobs withakeyeasily.
(33)	This doos opens easily with a key.
Additionally,ithas been observedby Fellbaum (1986) that the AdvP cannot precede the verb in themiddleconstruction (34), while suchword order is possible in a non-middle anticausative structure (35):
(34)	ThisPook(*easily) seads(easily).
(35)	Yes9esday,9he doos (quickly) opened (quickly).
Schäfer (2007) claims that the difference in the potential placement of the AdvP follows from the fact that (34)is structurally richer than (35) (that is, itcontains VoiceP) and from the additional assumption thatthe verbmovestothe Voicehead, as in (36a). By contrast, the anticausative sentence in(35)lacks the Voicelayer,sothe verbremainsin situ(36b),andtheadverbis able toprecede it.
Inthissense,asentencelike *ThisbookREADS isstillbadwithouttheadverbbecause placingfocus on reads leads to trivial informationgettingmappedintothe nuclearscope,asallbooksaresuch thatonecanread them, whereas not all dresses are such that one can button them (Marelj 2004).
54 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
(36)	a. [tp this book [t' T [voiceP [voice' Voice+reads [vp easily \ b. [tp this door [t' T [vp quickly [vp opened the door]]]'
(Thestructuresin(36) aremodelledafter Schafer2007.)
This factcanonlybeaccountedforif theadverbis mergedlow inthestructure asa VPadjunct, so before MiddleOp evenenters the syntacticderivation. Whatfollows,then,isthatthe adverb is not a semanticrequirement of the operator itself. Itis, however, needed ina sentence like This bookreads easily by the generic quantifierbecauseofthe well-formedness conditionon its bipartite scope structure (i.e., the nuclear scope mustn't be empty and must contain non-trivial information, cf. footnote 5). Crucially,this requirement on part of the quantifier is not built into the syntactic component,as predictedbytherevised semanticsof MiddleOpin (28), but holds atLF, giventhat it can besatisfiedin middles thatlack adverbs (examples (29)-(30)) by non-semantic means (i.e., focus placement).
4 The Slovenian ^-Construction
In this section, I turn to some of the main grammatical characteristics of the Slovenian middle construction and contrast them with those of the corresponding English construction. Recall from Section 1 that Slovenian middles differ from English middles in two key ways. First, they contain an additional syntactic category in the form of the reflexive morpheme se (37).6 Second, they allow a non-generic interpretation (38), which is not admissible in the English construction (39).
(37)	Ta knjiga se bere	zlahka. this book.NoM refl read.3sG	easily "This book reads easily."
(38)	Ta knjiga	se je vceraj brala.
this book.NOM.FEM refl aux.3sg yesterday read.L-PTCPL.FEM "Yesterday, this book was read."
(39)	*Yesterday, this book read.
Intended: "Yesterday, someone read this book."
The general intuition in the formal literature is that se is the source of cross-linguistic differences; as Alexiadou (2010, 3) succinctly proposes, "the morphology we see in the alternation should be taken seriously and is the device that helps us explain why [superficial valency reduction] is freer in some languages than others". Hence, it is the lack of se in English middles that limits their distribution in comparison to those in Slovenian.
6 Note that sentences (37) and (38) are similar to the Slovenian impersonal se-construction, in which the external argument DP (denoting the reader of the book) is also superficially absent from the overt structural representation, as in the following example, which is interpretatively equivalent to (37). (i)	To knjigo se zlahka bere.
this book.ACC refl easily read. "This book reads easily."
Rivero and Milojevic Sheppard (2003) argue at length that an impersonal sentence such as (i) is a two-argument structure; in addition to the overt internal argument DP (To knjigo), the sentence contains a thematic agent argument, which is realised as a phonologically null pronoun. By contrast, sentences (37) and (38) contain only one syntactic thematic argument (i.e., the internal one), whereas the agent is present only semantically as an existentially bound agent variable; cf. section 4.2 below.
MME1	LANGUAGE 55
4.1 The Syntax of se
As regards the syntactic role of se, I will adopt a recent proposal by Alexiadou, Anagnostopoulou, and Schäfer (2015, Chapter 4) and Schäfer (2017), in which a reflexive morpheme like se is analysed as a nominal DP that occupies the specifier position of a VoiceP, as in (40).7
(40) The VoiceP of a stf-sentence (Schäfer 2017)
Thecoresyntacticidea underlying(40) isthata sentencewith se is builtupinthe samemanner as a prototypical active transitive construction in terms of argument structure, in the sense that it containsbothan internal argument in the VP, knjiga,as wellasan external argument in the specifier of VoiceP; that is, se. At first look, the only thing that distinguishes a se-sentence like (38), Ta knjiga seje včeraj brala, from aprototypical active transitivesentencesuchas Marija je
J <■> NOM J	J	r	J *	J NOM J
včerajbralatoknjigocCC ("Yesterday,Marywas reading thisbook") is thedistributionofcase, in the sensethatthe internal argumentisassignednominativecase in the se-sentence whileaccusative case appears to be absent. Otherwise, the syntactic spine is the same in both sentences in the sense that either se or a referential DP such as Marija is merged as the specifier of a VoiceP.
Schäfers (2017, 5-6) account, however, posits that accusative case is actually presentin the structure, and that itis assignedto se. Sucha syntactic state of affairs arises due the fact that se istaken to be a minimalpronounthat iscompletelyunspecified for ^-features, in contrast withafully-fledged referentialDP suchas Marija inthe ordinarytransitivevariant. SinceSchäfer (zbd.)takes a set of valuedandinterpretable ^-features tobe aprerequisiteforaDP tobeassignednominativecase, the probethattriggers syntactic agreementskips se, which lacks suchfeatures,and assigns nominative totheotherreferential DP;that is, knjiga. By contrast,Schäfertakes accusative casetobe a type of dependent case that is assigned to a DP "if a different DP has valued the accessible phase head via AGREE"(2017, 6). Thismeans thatsincetheDP knjiga hasbeenassignednominativecase by the regularprocessofsyntactic agreement, se surfaceswith dependent accusative case.
Itake it thatsuchanapproach tothepositioningof se andtheconsequentinverteddistribution ofstructuralcaseisconceptually ontheright trackfor Slovenian se-sentences, sinceit captures twoveryimportantandquitegeneral structural factsabout theconstruction.Ontheone hand, itpredictswhy,intermsof themorphologicalmarkingonthe main verb, se-sentences pattern with ordinary active sentences. This is shown by the fact that a se-sentence realises the main verb as an l-participleina past tenseconstruction, asin (38). The reverseverbalmorphology holdsofcanonicalSlovenian passives,where themain verbis an n-participle complementing the passiveauxiliary biti ("be") inthe pasttense, as seen in thesentence Knjiga^M je bila bra«a,whichistheclosestmorphosyntactic equivalent of the English sentence Thebook was
7 Note thatthe syntacticderivation in(40)wasoriginallyproposedbyAlexiadou,Anagnostopoulou,andSchäfer(2015) and Schäfer(2017)forthemorphemes sich (in German) and se/si (inRomancelanguages).
56 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
being read. According to Embick (2004, 150), a verb surfaces with passive morphology only if it lacks an overt external argument. A se-sentence, however, does contain such an overt external argument, namely, se, hence its active morphology. On the other hand, this approach provides a straightforward explanation as to why the reflexive morpheme surfaces as se, and not for instance as dative si. That is, the morphophonological form is such precisely because se carries (dependent) accusative case.
4.2 The Semantics of se
The present contribution is tied to the semantics of se. First recall from Section 2.3 that the phono logically null MiddleOp is po sited as quite a complex semsuiticoperator — it introdu ces existentialquantificationover the apeni variable as well ^ ^neakqutntificarion ovrrsiiuaiicns. Thas ex^rins whu an Enghsh osiddk sentence isiü-formed tfit itnot iblerpeeted nrnorically, as in (39). By contrast, the corresponding Slovenian ^-sentence is perfectly well-formed with a non-generic (i.t.,euentive) interpreiuiirn (U8),so ti orems that seitlunciionallaaaimulef but overdvatiantofMiddleOpihat isnoeriodto agenericqcaniifie^ tsin(4e).
(41)	Ice] = Xf<e,<s,i».Xe<s>.3x[f(x, a)]
The semantics in (41) precisely illustrate Grahek's (2008, 52) otherwise quite informal observation that se reducestheexternalargument byoperatingon apredicate'sargumentgridin thesyntax.8 Concretely under the presentproposal, suchargumentreduction simplymeansthat se targets a syntactic structure that containsanopen agent variable introduced by Voice, and maps its denotation onto a scope in which itis existentially bound. (In terms of the 0-Criterion, this meansthat se saturates the external 0-roleinsemantics and thus precludes its assignment to an external argumentDP insyntax.)
Evidencefor (41) lies inthefactthat Slovenian se-sentences blockPPs denoting theexternal argument. While a canonical Slovenianpassiveallows its external argument tobe expressed via an od-phrase (42a) or a s-strani (42b) phrase, both PPs are ungrammatical in se-sentences (43), although an implicit agent is present in the semantic representation of both constructions.
(42)	a. S)q ja bil paivaUjaq ad acata.
cou.NOM Aux be pseisa.N-PTCPL by feidas "Ode sou wespseisadby disfeidas" b. Tazadava ja bila za epeazaqa e eteaqi majaqa pnaeCaeqika. idis issua.NOM aux be elsaedy seisa.N-PTCPL aq pant my peaeacaeeae "TCieieeua wae aleaady eaieae by my peaeacaeeae."
(Examples in (42) taken from Grahek (2008))
8 In relation to the semantics of se, the current proposal is different from Schäfers (2016, 14), which posits that se does not make a semantic contribution and that the agent variable, which is introduced by a special type of Voice head that Schäfer calls "E-bound Voice" (ibid.), is existentially bound from the very start. However, this would mean that the syntactic structure does not contain an open agent variable, which on the present account is problematic for the fact that Slovenian se-sentences with causative predicates allow the with-phrase (example (7)), which is licensed if the agent is initially an open variable in the syntactic-semantic representation.
L\NGUUGE 57
(43)	a. Hiša se je gradila (*od delavcev).
house.NOM refl Aux build.L-PTCPL (by workers)
"The house was being built (*by workers.)"
b. Jagode	so se jedle	(*s strani mojega sina).
strawberries.NOM aux refl eat.L-PTCPL (on part my son)
"Strawberries were being eaten (*by my son)."
The phrases are inadmissible in the two se-sentences in (43) because se, just like the null operator in English middles, requires as its argument an unsaturated Voice projection of type (e,(s,t)), whose open agent variable it closes off with an existential quantifier (41). An od-phrase (42a) and a s-strani phrase (42b), however, are semantically equivalent to the English by-phrase in that they intro duce the external argument into the structure, thereby closing the open agent variable by suphlyingthe d enotationoftheir complementDPs. Hence, admittingsuchp hrosesinto the structureofaar-sentencewouldturn thedenoSationoe the Voiceprojectioninto that oftype (s,tt, tid thus precOude thr merger of en
In addition, note that only anticausative verbs that are reflexively marked, such as odpreti ("open") in e4h),admit the Slavenian insteumental r-phríne, wtareaea verb like n0«st(hburss"), which is not reflexively marked, disallows it (45). On the present account, thecontrast arisesbecause sentence (44) contains an open agent variable, which is responsible for licensing the instrumental phraee and weich nuts closePe>íhbase. Conversd,, esn)enoe(45), ladinge)?, has anernantical^ vacuo^^icetuse, su ihe cve^aate isdianlowed(ccoseqhrntly, she possidle(nferen3e yf an agtnt	sOc CsUooo orlyinftes pragtndticaUb ocf. SihnOr ^dV^mco 2016)).
(44)	Vrata se zlahka odprejo (s ključem). doors REFL easily open (with key).
"These doors open easily with a key."
(454 Ta balon ( *se) zlahkapoči (*s škarjami). this bdloon oo-nR easiSy popo (with tciesoos). "This balls on popseadilysiwith scissors)"
Let's ncwTem!!! to the intra-hngjmsik pedoertiesof SWernan .yeteeoleofes oudinad d the beginning of this section. Under the present proposal, the generic se-sentence in (37), Ta knjiga se bere nlahkal differofrom ehe evontivosentence(n(38) , Co kdoigaseue včerojbeare, uninnimir^ally. In hothcasct,.ye fitst e3lstentileiy brndstheopen3gentvhri ao^lr^, rcudling in shefaet dint the dendution nfVoiceP (40) is a function in which only the event variable remains open, as in (46).
(4(54 iVoieeP] = Xe<s>.3x[AGENT(x, e) a READ(e) a PATlENT(the book, e)]
The semantic difference between sentences (37) and (38) arises because, in the former sentence, a generic quantifier,which is available as anindependent semanticoperatorinall languagesand is prototypically triggered when the sentence refers to an unbounded set of events temporally located inthepresentdomain (Chierchia 1995),isintroduced to saturate the remaining open
58 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
event variable.9 Applied to (46), the generic quantifier brings about the by now well-known middle LF, paraphrased as Generally, when (an arbitrary) someone reads the book, she does so easily.
4.3 The Covert English Middle Operator and the Slovenian se
As a final note, let's consider the fact that English middles admit fewer verb types than the corresponding Slovenian se-sentences. Perhaps most strikingly, an English middle sentence is ill-formed with a stative verb (47), which is otherwise admissible in the corresponding Slovenian variant (48).
(47)	*This castle sees easily from afar.
Intended: "Anyone can see this castle with ease from afar."
(48)	Ta grad se zlahka vidi od daleč. this castle.NOM refl easily see from afar "Anyone can see this castle with ease from afar."
The fact that (48) is a well-formed sentence should come as no surprise on the present account, as its grammaticality simply follows from the general sentence composition rules that have been proposed for Slovenian se-sentences. Concretely, the VP videti grad "see castle" merges with a Voice head introducing the external participant variable (in this case, the experiencer), which is then saturated when se is subsequently introduced as the specifier of VoiceP. The reason as to why (47) is ill-formed, however, is quite mysterious, especially given that MiddleOp is posited as an operator that simply targets a syntactic structure with an open variable denoting the external participant.
I am unable to offer a precise explanation of this fact, but conclude on a more informal and speculative note by tentatively proposing that the difference in grammaticality between (47) and (48) is due to the fact that the English lexicon lacks a lexical item such as se, which is a nominal pronoun. In other words, English middles are formed by an operator that is not lexicalised as an independent nominal expression, while the operator itself plays a more complex semantic role than se, invariably introducing generic quantification over the event variable. In short, because MiddleOp in English middles is a more complex operator than se in Slovenian sentences, it is also more limited in its application.10
9	For the eventive variant (38), I assume that existential closure is applied to the remaining event variable at TP, so the sentence is interpreted as There was an event (temporally included in the past time interval denoted by yesterday) in which someone was reading the book (Dowty 1982).
10	In this respect, an anonymous reviewer questions why MiddleOp is, on the present account, taken to be a different syntactic constituent (i.e., the head of its own phrase above VoiceP) from the reflexive morpheme se, which is posited as the specifier ofVoiceP. The reviewer claims that such an analysis potentially misses a cross-linguistic generalisation about the grammatical properties that English middles and Slovenian se-sentences have in common. I would like to emphasise that this is precisely the point — the English middle construction and the Slovenian se-construction are syntactically different beasts. English middles, lacking a nominal element like se, are structurally deficient, which is why their well-formedness seems to be dependent on genericity (according to Hartl (2012), genericity obviates the requirement that a transitive verb like read combine with an external argument, which is shown by the fact that an eventive sentence like **This book read yesterday is bad precisely because no such argument is present in its structure) and why they cannot be formed from just about any predicate type (example (47)), whereas the formation of se-sentences is unconstrained (example (48)). In other words, the present proposal highlights the fact that the equivalent interpretation of Slovenian generic se-sentences and English middles arises due to different — albeit — similar underlying operations (i.e., MiddleOp binds the agent and event variables in one fell swoop, while se only binds the agent variable). Furthermore, note that this is consistent with Lekakou's (2004) cross-linguistic observation that different languages may not employ the same grammatical strategies to form middles.
MME1	LANGUAGE 59
5 Conclusion
This paper has discussed the formation of English and Slovenian middle sentences from the perspective of a minimalist syntactic approach to argument structure. The paper has first presented the conceptual and empirical advantages of Bruening's proposal (2012), which posits that English middles are derived via a null operator that requires an unsaturated Voice projection, showing how his approach elegantly (i.e., without additional stipulations) accounts for certain key properties of the construction, such as the arbitrary reference of the agent. In addition, it has been proposed that the adverb is not a semantic argument of the null middle operator. Lastly, the paper has discussed the reflexive morpheme se in Slovenian se-sentences, and proposed that it has a semantics similar to that of the null English operator.
References
Ackema, Peter, and Maaike Schoorlemmer. 1994. "The Middle Construction and the Syntax-Semantics Interface." Lingua 93 (1): 59-90. https://doi.org/10.1016/0024-384l(94)90353-0.
Alexiadou, Artemis. 2010. "On the Morphosyntax of (Anti)-Causative Verbs." In Lexical Semantics, Syntax, and Event Structure, edited by Malka Rappaport Hovav, Edit Doron, and Ivy Sichel, 177-203. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199544325.003.0009.
Alexiadou, Artemis, Elena Anagnostopoulou, and Florian Schäfer. 2015. External Arguments in Transitivity Alternations: A Layering Approach. Oxford: Oxford University Press.
Borer, Hagit. 2004. "The Grammar Machine." In The Unaccusativity Puzzle, edited by Artemis Alexiadou, Elena Anagnostopoulou, and Martin Everaert, 288-331. Oxford: Oxford University Press. https://doi. org/10.1093/acprof:oso/9780199257652.003.0012.
Bruening, Benjamin. 2012. "By Phrases in Passives and Nominals." Syntax 16 (1): 1-41. https://doi. org/10.1111/j.1467-9612.2012.00171.x.
Chierchia, Gennaro. 1995. Dynamics of Meaning. Chicago: University of Chicago Press.
Chomsky, Noam. 2001. "Derivation by Phase." In Ken Hale: A Life in Language, edited by Michael Kenstowicz, 1-52. Cambridge, MA: MIT Press.
Dowty, Don. 1982. "Tense, Time Adverbs and Compositional Semantic Theory." Linguistics and Philosophy 5: 23-55.
Embick, David. 2004. "Unaccusative Syntax and Verbal Alternations." In The Unaccusativity Puzzle, edited by Artemis Alexiadou, Elena Anagnostopoulou, and Martin Everaert, 137-58. Oxford: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/9780199257652.003.0006.
Fagan, Sarah M. B. 1992. The Syntax and Semantics ofMiddle Constructions: A Study with Special Reference to German. Cambridge: Cambridge University Press.
Fellbaum, Christiane. 1986. On the Middle Construction in English. Bloomington, Indiana: Indiana University Linguistics Club.
Grahek, Sabina. 2008. "Middles in Slovene." Leeds Working Papers in Linguistics and Phonetics 13: 44-75.
Härtl, Holden. 2012. "Generic Rescue: Argument Alternations and the Monotonicity Condition." In Repairs: The Added Value of Being Wrong, edited by Patrick Brandt and Eric Fuß, 95-130. Berlin: De Gruyter.
Klingvall, Eva. 2015. "A Secret Agent in the Middle?" Working Papers in Linguistics (Department of English in Lund) 5: 91-114. Accessed August 23, 2018. http://www.sol.lu.se/fileadmin/media/forskning/ workingpapers/engelska/vol05/klingvall-wp05.pdf.
60 Jakob Lenardic The Middle Construction in Minimalist Syntax: English and Slovenian
E L O P
Kratzer, Angelika. 1996. "Severing the External Argument from Its Verb." In Phrase Structure and the Lexicon, edited by Johan Rooryck and Laurie Zaring, 109—37. Dordrecht: Kluwer Academic Publishers. https:// doi.org/10.1007/978-94-015-8617-7_5.
Krifka, Manfred, Francis Jeffry Pelletier, Gregory N. Carlson, Alice ter Meulen, Godehard Link, and Gennaro Chierchia. 1995. "Genericity: An Introduction." In The Generic Book, edited by Gregory N. Carlson and Francis Jeffry Pelletier, 1—124. Chicago: University of Chicago Press.
Legate, Julie Anne. 2014. Voice and v — Lessons from Acehnese. Massachusetts, MA: MIT Press.
Lekakou, Marika. 2004. "Middles as Disposition Ascriptions." In Arbeitspapier Nr. 117: Proceedings of the Conference "Sub8 — Sinn und Bedeutung", edited by Cécile Meier and Matthias Weisgerber, 181—196. Konstanz: Universität Konstanz.
Levin, Beth, and Malka Rappaport Hovav. 1995. Unaccusativity: At the Syntax-Lexical Semantics Interface. Cambridge, MA: MIT Press.
Marantz, Alec. 1997. "No Escape from Syntax." University of Pennsylvania Working Papers in Linguistics 4: 201-25.
Marelj, Marijana. 2004. Middles and Argument Structure Across Languages. Utrecht: LOT Publishing.
McConnell-Ginet, Sally. 1994. "On the Non-Optionality of Certain Modifiers." In Proceedings of SALT4, edited by Mark Harvey and Lynn Santelmann, 230-50. https://doi.org/10.3765/salt.v4i0.2458.
Ramchand, Gillian. 2008. Verb Meaning and the Lexicon: A First Phase Syntax. Cambridge: Cambridge University Press.
Rivero, Maria-Luisa, and Milena Milojevic Sheppard. 2003. "Indefinite Reflexive Clitics in
Slavic: Polish and Slovenian." Natural Language & Linguistic Theory 21: 89-155. https://doi. org/10.1023/A:1021841517604.
Schäfer, Florian. 2007. "Middles as Voiced Anticausatives." In Proceedings of NELS 37 (Volume 2), edited by Emily Elfner and Martin Walkow, 183-96. Accessed August 31, 2018. http://ling.auf.net/ lingbuzz/000400.
—. 2016. "Medio-Passives within a Formal Typology of Voice." Manuscript. Accessed July 20, 2018. https:// amor.cms.hu-berlin.de/~schaeffl/papers/mediopassives2.pdf.
—. 2017. "Romance and Greek Medio-Passives and the Typology of Voice." In The Verbal Domain, edited by Roberta D'Alessandro, Irene Franco, and Angel J. Gallego, 129-52. Oxford: Oxford University Press. https://doi.org/10.1093/oso/9780198767886.003.0006.
Schäfer, Florian, and Margot Vivanco. 2016. "Anticausatives are Weak Scalar Expressions, Not Reflexive Expressions." Glossa: A Journal of General Linguistics 1 (1): 1-36. http://doi.org/10.5334/gjgl.36.
MME1	LANGUAGE 61
E L O P
Frančiška Lipovšek	2018' VoL 15 63-78/(!20)
TT . . ri. ... -C, .	revije.ff.uni-lj.si/elope
University of Lpbljam^bwma	doi: io.4312/elope.15.2.63-78
UDC: [811.111'367.625:81'373.612.2]:811.163.6
The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene Fictive Motion Expressions
ABSTRACT
The paper presents a study on fictive motion (FM) uses of motion verbs in English and Slovene from the point of view of conveying path-related information. An FM expression describes a static scene in terms of motion (e.g. The road weaves through a range of hills). Motion verbs in FM uses do not describe actual motion events, but may refer to certain properties of the path by virtue of their meanings. English and Slovene FM expressions exhibit different behaviours in this respect. Many English verbs display meaning components that can be metonymically mapped onto the properties of the path. The meanings of Slovene verbs are less specific, so that such properties need to be expressed verb-externally in Slovene FM expressions.
Keywords: fictive motion; motion verbs; manner of motion; path; metonymy; metonymic mapping
Vloga glagolov premikanja pri opisu poti v angleških in slovenskih izrazih fiktivnega gibanja
POVZETEK
V članku je predstavljena študija glagolov premikanja v angleških in slovenskih izrazih fiktivnega gibanja z vidika informacij, ki jih ti glagoli posredujejo o poti premikanja. Izraz fiktivnega gibanja opisuje statični prizor na podlagi gibanja oziroma premikanja (npr. Cesta vijuga med hribi). Glagoli premikanja v tovrstni rabi ne izražajo dejanskega premikanja, vendar se lahko na podlagi svojega pomena nanašajo na nekatere vidike poti. Angleški in slovenski izrazi fiktivnega gibanja se v tem pogledu vedejo različno. Mnogo angleških glagolov vsebuje pomenske komponente, ki omogočajo metonimično preslikavo na določene lastnosti poti. Pomeni slovenskih glagolov so manj specifični, zaradi česar se v slovenskih izrazih fiktivnega gibanja za opis poti uporabljajo druga, tj. neglagolska sredstva.
Ključne besede: fiktivno gibanje; glagoli premikanja; način premikanja; pot; metonimija; metonimična preslikava
MME1
LANGUAGE 63
The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene Fictive Motion Expressions
1	Introduction
Fictive motion1 is a construal of a static scene in terms of motion. Sentence (1a) below communicates a stationary situation by describing a scene that, owing to the verb lead, invites motion: our eyes mentally scan an imaginary path leading onto the terrace. Similarly, there is an imaginary entity moving along the road in (1b).
(1) a) The French windows lead out onto a wide shady terrace. (CALD) b) The road weaves through a range of hills. (OALD)
Sentence (1b) contains a motion verb commonly found in real (i.e. actual) motion sentences like He hurried on, weaving through the crowd (OALD). Nevertheless, weave in (1b) cannot express motion since a static scene with a stationary trajector (TR)2 is described. Rather, it provides information about the TR: the manner component of its meaning is mapped onto the shape of the road.
The paper presents a study focusing on uses exemplified by weave, above. It examines the use of motion verbs conveying path-related information3 in English and Slovene FM expressions. The study is presented in the second half of the paper. The two introductory sections are devoted to the theoretical aspects of fictive motion and its linguistic manifestation.
2	Understanding Fictive Motion
The processing of FM sentences involves "mentally simulated motion" (Matlock 2004a, 2004b): the conceptualizer (i.e. the language user) mentally simulates visual scanning in (1a) and movement in (1b). Motion is imposed on the scene subjectively; the conceptualizer is neither part of the objective scene nor moving through it (Langacker 2000, 6).
As observed by Talmy (2000, 104), "languages systematically and extensively refer to stationary circumstances with forms and constructions whose basic reference is motion." Representations of stationariness in terms of motion are much more common than representations of motion in terms of stationariness — an asymmetry reflecting a "cognitive bias toward dynamism" (2000, 171).
In his ground-breaking work on nonveridical phenomena, Talmy (2000) subsumes fictive motion under "general fictivity", a cognitive pattern characterized by "discrepancy between two different cognitive representations of the same entity, where one of the representations is assessed as being more veridical than the other" (Talmy 2000, 100). The discrepant representations as
Also referred to as abstract/subjective/virtual motion.
The carrier of the relation in a spatial scene; the entity (stationary or in motion) whose location is described in relation to the landmark (LM), a reference point entity (cf. Langacker 1987, 217ff.).
Path-related information is used as a cover term to refer to the properties of the TR in a static scene. The path — TR connection is obvious, since the TR either coincides with or in some way accommodates the imaginary path travelled by a fictively moving entity.
64 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
E L O P
well as the assessment of their veridicality are produced by distinct cognitive subsystems within the cognition of a single individual. Talmy proposes the term "factive" to designate the more veridical representation and the term "fictive" to designate the less veridical one, pointing out that the ascribed degree of veridicality is the result of a cognitive assessment rather than a reflection of objective (un)reality.
Fictive motion is a manifestation of "fictive X" (Talmy 2000, 101), a category comprising any possible fictive representation (fictive presence, absence, change, etc.) within the general fictivity pattern. There are several dimensions along which discrepancies of the factive/fictive opposition can be identified. Relevant to the present study is the "state of motion", where the more veridical, factive representation involves stationariness and the less veridical, fictive one motion. The former is based on belief and the latter on linguistic form. For illustration, our knowledge of the world tells us that sentence (1b) describes a factively static scene; it is our belief that roads do not move in space-time. Nevertheless, the presence of a motion verb together with a directional PP evokes a fictive, nonveridical representation of something in motion. The sentence exhibits a "linguistic pattern in which the literal meaning of a sentence ascribes motion to a referent that one otherwise normally believes to be stationary" (Talmy 2000, 101). The degree to which motion is experienced may vary across speakers.
In line with the view that language mirrors human conceptualization and that cognitive processes are largely based on the principles ofvisual perception (cf. Jackendoff 1994), Talmy (2000, 102f.) points out the analogy between linguistic and visual manifestations of fictive motion. If a visual scene allows two simultaneous representations, the one perceived in terms of stationariness will be assessed as the more palpable, factive one, and the one perceived in terms of motion as the less palpable, fictive one. Thus, a sense of motion is ascribed to a scene that one otherwise normally sees to be stationary. There is an obvious parallel between the believing/inferring4 opposition characterizing fictive-motion sentences, and the seeing/sensing opposition characterizing visual scenes.
Fictive motion is also interpreted in terms of conceptual blending (or conceptual integration), a cognitive operation of integrating two or more input mental spaces5 into a new, blended space with an emergent structure of its own (Fauconnier 1997; Fauconnier and Turner 2002). Mapping between the elements of the input spaces makes it possible to convey immobility and motion at the same time. For illustration, the subject (the road) in (1b) above evokes an immobility space and the predicate (weaves through a range ofhills) a motion space. The elements of the immobility space thus include a stationary TR (the road). The elements of the motion space include a moving TR and its trajectory (i.e. the path along which it moves). Cross-space mapping connects the trajectory from the motion space to the road from the immobility space. The trajectory and the road project onto the same single element in the blended space: the stationary TR emerges as a path designated for a moving TR.
According to Jiménez Martínez-Losa (2007, 564f.), the conceptualization of fictive motion relies on the motion metaphor, a conceptual metaphoric shift between a factive motion event as the source domain and a fictive motion event as the target domain. Jiménez Martínez-Losa notes that fictive motion events have no parallel in reality, and are therefore more abstract than factive motion events. The direction of the shift, namely from factive motion to fictive motion, thus accords with the fact that metaphor typically involves the mapping of elements belonging to a
4	Used in the sense 'inferring the meaning as determined by the language system'.
5	For the theory of mental spaces see Fauconnier (1994).
MME1	LANGUAGE 
65
concrete domain onto elements belonging to an abstract domain. Moreover, the trajector of a fictive motion event is conceptualized as a moving entity, i.e. like the trajector of a factive motion event, which accords with the view that metaphor is based on the identity of relations (cf. Lakoff 1987; Lakoff and Johnson 2003). Last but not least, Jiménez Martínez-Losa points out that the conceptualization of fictive motion foregrounds the path that the trajector of a factive motion event would follow (e.g. through a range of hills in (1b)). She argues for an underlying metonymic relationship where the motion along the path activates the path of motion (motion along a path for path metonymy), or rather, where the conceptualization of a path naturally results from the action of motion along the path (action for result metonymy) (ibid., 565f.). In sum, the conceptualization of fictive motion involves the motion metaphor motivated by the action for result metonymy.
Recognizing metonymy in fictive motion is of particular relevance to our paper, because metonymic relations can be observed in FM uses of motion verbs, especially those expressing the manner of motion. Matlock (2004b, 11—12) discusses crawl, race and stagger from the point of view of information they can convey about the TR in a static scene (representing the path in the fictive representation). She argues that a highway that crawls through the city is construed differently from a highway that races through the city, the former being associated with slow motion and the latter with fast motion. Similarly, a footpath that staggers from point A to point B is likely to be conceptualized in terms of several sharp bends, associated with the particular way of walking denoted by stagger. Such metonymic inferences based on associations demonstrate that metonymy is more than just a "stand-for" relation (cf. Barcelona 2011; Langacker 2000; Panther and Thornburg 2005).
3 FM Expressions
Fictive motion manifests itself in language in several ways. Talmy (2000, 105) identifies a set of conceptual features whose values, or rather combinations of these values, distinguish different categories of fictive motion. For illustration, the FM construals below differ with regard to three principal features: +/- obligatoriness of factive motion, +/- observer-based fictive effect, +/-factiveness of the fictively moving entity.
(2) a) The fence goes from the plateau to the valley. (Talmy 2000, 138)
b) I sat in the car and watched the scenery rush past me. (Talmy 2000, 132)
In (2a), no element needs to actually move, the fictive effect does not depend on the observer (i.e. it is observer-neutral), and the fictively moving entity exists only in the observer's imagination (i.e. it is fictive). In (2b), a factively moving observer is crucial for the fictive effect, the fictive effect is obviously observer-based, and the fictively moving entity exists in the real world (i.e. it is factive).
According to Talmy's classification of FM categories, the two FM expressions under (2) represent a "coextension path" (Talmy 2000, 138f.) and "frame-relative motion" (Talmy 2000,130— 34) respectively. With few exceptions, the categories are referred to as "paths" (coextension paths, orientation paths, radiation paths, access paths, etc.), reflecting the fact that the path is foregrounded in fictive motion and that an FM expression essentially depicts the (factive, stationary) TR — LM arrangement in terms of a path. For example, in a coextension path (e.g. (2a)) the path runs along the TR's extent whereas in an access path (e.g. The bakery is across the street from the bank (Talmy 2000, 137)), the TR represents the endpoint on the path.
66 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
E L O P
We will focus on coextension paths because we are interested in the relation between the manner of motion and the properties of the factive TR. If the manner of motion evokes information about the path, it is reasonable to limit the study to cases where properties of the path can be understood as properties of the TR. This condition is met by coextension paths, which involve a path moving along/over/across a spatially extended object (TR).
A coextension path is defined as a "depiction of form, orientation, or location of a spatially extended object in terms of a path over the object's extent" (Talmy 2000, 138). The path is not necessarily linear, and exactly what is imagined as moving along the path is not specified: Talmy points out it can be an observer,6 the focus of one's attention,7 or the object itself.8 This goes hand in hand with Matlock's (2004a, 1396) observation that the processing of an FM sentence combines several types of simulations. Nevertheless, although these simulations are not mutually exclusive, they are not equally salient. For example, fictive motion along an object that is not normally perceived as travelable by humans (e.g. a fence, as opposed to a road) is more likely to be simulated in terms of sequential mental scanning than in terms of an imagined traveller (cf. also Rojo and Valenzuela 2010, 255).
TRs in coextension path expressions are spatially extended entities that need not be inherently long, but must be at least conceptually extendable (Matlock 2004b, 7). For illustration, the sentences below differ in acceptability because the TR in (3b) does not readily admit of a conceptualization in terms of linear extension.
(3)	a) The table goes from the kitchen wall to the sliding glass door. (Matlock 2004b, 7) b) ?? The hoola hoop runs from the door to the couch. (Matlock 2004b, 8)
The TR is usually associated with linearity and horizontal extension; nevertheless, certain types of TRs are typically associated with vertical extension (e.g. tall buildings and landscape features like mountains and cliffs). Moreover, some TRs can be construed in different ways through different types of verbs (Egorova, Tenbrink, and Purves 2018, 14). Egorova, Tenbrink, and Purves present mountain ridges as a case in point, demonstrating that they can be construed as being linear and having a complex shape (4a), as horizontally extended without a complex shape (4b), or as vertically extended (4c):
(4)	a) A slender scimitar-shaped ridge curved upwards.
b)	To the north an almost horizontal ridge runs west over Whymper, Croz ...
c)	Glancing above, the ridge soared in a series of ice towers.
(Egorova, Tenbrink, and Purves 2018, 14)
Each of the above construals arises from the semantics of a different type of verb. The verb curve encodes a change in direction; the verb run encodes motion along a TR of non-specified shape; the verb soar encodes upward vertical motion combined with manner ('rise quickly'). The manner of motion generally conveyed by run appears to be suppressed in fictive motion (cf. Matsumoto 1996;
6	Cf. the local path/frame (Matsumoto 1996, Talmy 2000). With a moving observer, the scope of attention is a "local" one: the observer's immediate field of view changes all the time.
7	Cf. the global path/frame (Matsumoto 1996, Talmy 2000). With a stationary observer, the scope of attention is a "global" one: the observer moves the focus of attention by visually scanning along the path, but his/her field of view remains unchanged.
8	Imagined as if "advancing along its own axis" (Talmy 2000: 138).
MME1	LANGUAGE 67
see also the discussion below). The most relevant to our study are FM expressions like (4c) because the manner-of-motion component of the verb's meaning indirectly conveys information about some property of the TR: with the verb soar, the ridge is construed as ascending at a steep angle.
In terms of syntax, a typical coextension path sentence displays the structure S + P + A:
(5)	This road goes to Birmingham. (CALD)
The subject represents the TR and the adjunct the LM. The adjunct is realized by a PP expressing direction. The predicator contains a motion verb. It should be noted at this point that none of the constituents corresponds to a trajectory (i.e. path) since the sentence (factively) describes a static scene. It is only the fictive representation of the sentence that involves motion. There, the subject NP is conceived of as the trajectory (or the entity along/over which the trajectory runs) of an imaginary TR that has no syntactic realization.
Needless to say, there are variations on the basic pattern (cf. Matlock 2004b). For example, (6a) below displays a set of two LMs. In (6b), the LM is represented by a direct object. The LM remains implicit in (6c), where the verb is followed by a directional particle and an adjunct of extent. In (6d), the verb occurs without any complementation because the relative orientations of the distributaries are inferable from its lexical meaning.
(6)	a) A huge crack went from the top to the bottom of the wall. (CALD)
b)	Turn left where the lane meets the main road. (CALD)
c)	The tree's roots go down three metres. (CALD)
d)	A sample of water was taken from the point where the river bifurcates. (CALD)
Last but not least, the verb in the predicator need not be an actual motion verb (see (6b) and (6d) above), or, as noted by Ruppenhofer (2006, 310, fn. 4), "not all fictive motion uses are productively formed from verbs of (actual) motion."9
In his seminal study on English and Japanese motion verbs, Matsumoto (1996) proposes two conditions that govern the use of FM expressions involving coextension paths. The conditions read as follows:
a.	The path condition: Some property of the path of motion must be expressed.
b.	The manner condition: No property of the manner of motion can be expressed unless it is used to represent some correlated property of the path.
(Matsumoto 1996, 194)
The path condition states that an FM expression is acceptable only if it includes information about the shape or some other aspect of the path. For illustration (adapted from Matsumoto 1996: 194):
(7)	a) The road began to ascend.
b)	*The road began to run.
c)	The road began to run straight.
9 Ruppenhofer mentions taper and jut as examples of "specifically fictive" verbs, i.e. non-motion verbs occurring specifically in FM expressions (e.g. Turn left where the road tapers (off) into a track. (CALD)).
68 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
E L O P
The path condition makes relevant the distinction between path-conflating verbs like ascend and manner-conflating (or manner-of-motion) verbs like run. In (7a), it is evident from the meaning of the verb that the road is positioned on a slope. Conversely, run in (7b) conveys no path-related information. Sentence (7c) is acceptable owing to an adverb whose meaning is correlated with the shape of the path.
The manner condition imposes restrictions on the use of manner-of-motion verbs in FM constructions. As the term implies, manner-of-motion verbs describe the mode in which an object moves along its path. Sometimes the manner of motion evokes a property that is relevant to the path (zigzagging, for example, evokes a Z-shape). The manner condition precludes any uses of manner-of-motion verbs where the verb cannot be interpreted as (indirectly) conveying some path-related information. For illustration (adapted from Matsumoto 1996, 194):
(8)	The road wanders/*walks through the park.
The verb wander can represent fictive motion because its meaning evokes some kind of irregular path shape corresponding to walking around without any clear purpose or direction. Conversely, no path feature is evoked by walk, rendering the sentence unacceptable.
A slight problem is now presented by (7c) above. The sentence is acceptable although the verb run does not evoke any path-related property:
(9)	The road began to run straight. (= 7c)
Matsumoto (1996, 200) proposes that the sentence is not in conflict with the manner condition because in the case of run, the semantic component of manner is suppressed in FM expressions. He also observes that run is the only manner-of-motion verb exhibiting this type of behaviour.10
The path and manner conditions introduced by Matsumoto (1996) represent a major contribution to the field and are extensively referenced in the literature. Walinski (2015) argues for a third condition, the so-called instrument condition stating that "no property of motion instrument can be expressed in a coextension path, unless it is used to represent some specifically correlated property of the path" (Walinski 2015, 98). The condition, supported by the results of a corpus-based study (Walinski 2015), precludes instrument-conflating motion verbs (i.e. motion verbs whose meanings entail an instrument of motion; e.g. drive, pedal, ride) from structuring fictive motion. For illustration (cf. Walinski 2015, 90):
(10)	This road goes/*drives to London.
Nevertheless, Walinski at the same time admits that since instrument and motion form a semantic cline, the instrument condition partly overlaps with Matsumoto's manner condition. As a matter of fact, Matsumoto (1996: 190) subsumes the type of vehicle used in motion under the aspects of manner.
4 The Aim of the Study and Methodology
The aim of the study was to examine the role of motion verbs in English and Slovene FM expressions from the point of view of their potential to convey path-related information. With
10 The same holds for its Japanese counterpart (ibid.). This, however, is not universal. See, for example, Takemoto (2010) for a comparison with French and German.
MME1	LANGUAGE 69
a view to identifying similarities/differences between the two languages, the following research questions were formulated:
i)	Which types of motion verbs are employed by English and Slovene FM expressions to convey path-related information?
ii)	Which properties can be expressed by these verbs and how?
Relevant FM expressions were searched in language corpora by using the Sketch Engine corpus tool. The goal was not to perform a quantitative analysis but to identify the scope of motion verbs in FM uses. The following corpora were used: the British National Corpus (BNC), the British English Corpus (ukWaC) and the English Web Corpus from the TenTen Corpus Family (enTenTen) for English, and the Slovenian Reference Corpus (FidaPLUS v2) and the Slovenian Web Corpus from the TenTen Corpus Family (slTenTen) for Slovene.
The corpora were queried for verbs that collocate with nouns representing TRs in coextension paths. The selection of nouns was partly based on Egorova, Tenbrink, and Purves' (2018) research into fictive motion in the context of mountaineering. Drawing upon their list of 10 most frequent landscape terms occurring in FM expressions (Egorova, Tenbrink, and Purves 2018, 13), the following four nouns were selected: mountain (Slo: gora) and slope (Slo: pobočje) as entities associated with vertical extension, ridge (Slo: greben) as an entity that can be associated with both vertical and horizontal extension, as well as with linearity and complex shape (see the previous section), and valley (Slo: dolina) as an entity associated with linearity and horizontal extension but also allowing a construal in terms of vertical extension (Egorova, Tenbrink, and Purves 2018, 15). Path (Slo: steza) and road (Slo: cesta) were added to the list as entities associated with linearity or a complex shape, and differing with regard to the mode of travel.
With five corpora used, there were five queries for each noun. Each individual query generated a list of verbal collocates arranged according to the collocation strength. The maximum number was set at 200 (relevant for nouns with a high number of collocates). The list was examined manually to extract the verbs whose meanings conflate motion with path/manner. The verbs whose meanings do not evoke any correlated path properties (e.g. run) and any uses representing FM categories other than coextension paths (e.g. surround used in an advent path (cf. Talmy 2000, 135) or poke in a demonstrative path (Talmy 2000, 109)) were not included. Several collocations had to be discarded because they did not represent the targeted subject + verb structures but complex NPs (e.g. ridge scrambling, road humps), or the noun was not used with the targeted meaning (e.g. English path, Slovene greben). For each language, a final list was compiled comprising all verbs from the individual lists (three for English, two for Slovene) that were recognized as relevant to the study.
5 Results
Among the collocations generated by the corpora search, the following noun + verb combinations were identified as representing FM uses of motion verbs conveying path-related information.
1. mountain (En) / gora (Slo) a) Vertical motion
English: climb, descend, dip, drop, fall, plunge, rise, sink, soar, tumble Slovene:11 dvigati/dvigovati/dvigniti se, spuščati se, vzdigovati se, vzpenjati se
11 The English translations are given in Table 1. 70 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene ..
b) Horizontal motion English: roll, sweep Slovene: -2. (mountain) ridge (En) / (gorski) greben (Slo)
a)	Vertical motion
English: ascend, climb, descend, drop, fall, rise
Slovene: dvigati/dvigovati/dvigniti se, spuščati/spustiti se, vzpeti se
b)	Horizontal motion English: curve, sweep
Slovene: obrniti se, valoviti, viti se
3. slope (En) / pobočje (Slo)
a)	vertical motion
English: ascend, climb, descend, drop, fall, plunge, rise
Slovene: dvigati/dvigovati se, padati, spuščati/spustiti se, vzpenjati se
b)	Horizontal motion
English: ease, roll, slip, sweep, turn, wind Slovene: -4. valley (En) / dolina (Slo)
a)	vertical motion
English: climb, descend, drop, fall, plunge, rise Slovene: dvigati/dvigniti se, spuščati/spustiti se
b)	Horizontal motion
English: flow, roll, sweep, turn, wind Slovene: viti se, zaviti
5.	path (En) / steza (Slo)
a)	vertical motion
English: ascend, climb, descend, dip, drop, fall, rise
Slovene: dvigati/dvigniti se, spuščati/spustiti se, vzpenjati/vzpeti se
b)	Horizontal motion
English: criss-cross, curve, ease, hop, loop, meander, roll, snake, sweep, swing, turn, twist, undulate, veer, wander, weave, wind, zigzag Slovene: cikcakati, obrniti se, vijugati, viti se, zaviti
6.	road (En) / cesta (Slo)
a) vertical motion
English: ascend, climb, descend, dip, drop, fall, rise
Slovene: dvigati/dvigovati/dvigniti se, povzpeti se, spuščati/spustiti se, vzpenjati/vzpeti se
MME1	LANGUAGE 71
b) Horizontal motion
English: criss-cross, curve, loop, meander, snake, sweep, swing, turn, twist, undulate, veer, weave, wind
Slovene: obrniti se, vijugati, viti se, zavijati/zaviti The data are summarized in the table below, with the verbs classified into five groups:
i)	directional verbs encoding upward vertical motion (e.g. rise)
ii)	directional verbs encoding downward vertical motion (e.g. fall)
iii)	directional verbs encoding a change of direction and/or motion along a curved path (e.g. curve)
iv)	manner-of-motion verbs encoding motion along a path of complex shape (e.g. zigzag)
v)	manner-of-motion verbs encoding motion along a path of non-specified shape (e.g. hop)
Table 1. Motion verbs conveying path-related information
	English	Slovene
Upward motion	ascend, climb, rise, soar	dvigati/dvigovati/dvigniti se ('rise'), povzpeti se ('ascend, climb'), vzdigovati se ('rise'), vzpenjati/vzpeti se ('ascend, climb')
Downward motion	descend, dip, drop, fall, plunge, sink, topple, tumble	padati ('fall'), spuščati/ spustiti se ('descend')
Change of direction, curved path	curve, loop, turn, veer, sweep, swing	obrniti se ('turn'), zavijati/ zaviti ('turn')
Manner + complex shape	criss-cross, meander, snake, twist, undulate, weave, wind, zigzag	cikcakati ('zigzag'), valoviti ('undulate'), vijugati ('weave'), viti se ('wind')
Manner + non-specified shape	ease, flow, hop, roll, slip, wander	--
6 Discussion
The results demonstrate a discrepancy between English and Slovene FM expressions in the use of motion verbs that convey path-related information. The English verbs outnumber the Slovene ones12 for two reasons. First, quite a few English verbs do not have distinct Slovene equivalents, even in actual motion uses, and share an equivalent with a relatively general meaning (e.g. meander, snake, twist, weave, wind —> viti se/vijugati). Second, some of them have Slovene equivalents that resist fictive motion uses (e.g. climb ^ plezati; plunge ^ planiti, strmoglaviti; topple ^ prevrniti se; hop ^ skakljati).
12 It should be noted that the Slovene set of upward motion verbs basically comprises lexical aspectual variants of only two verbs: dvigniti se 'rise' and vzpeti se 'ascend' (see Table 1).
72 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
E L O P
The inventory of English verbs includes verbs whose meanings display semantic components pertaining to some other aspects of motion. For illustration, the verb plunge designates a sudden movement. This component of meaning translates into a property of the path (steepness in our case; see the discussion below) that can only be expressed verb-externally in Slovene. Compare:
(11)	a) High mountains plunge into glittering lakes, creating almost fjord-like scenery.
(ukWaC)
b) Visoke gore [...] strmopadajo proti Tihemu oceanu. (FidaPLUS v2)
'High mountains [...] fall steeply towards the Pacific Ocean.'
The verbs presented in Table 1 primarily convey information about the direction of movement or the path-shape related manner of motion (the exception being verbs encoding motion along a path of non-specified shape, which will be treated separately). Nevertheless, as demonstrated by plunge above, some verbs at the same time encode information about speed. Among the verbs encoding direction or path shape, the following ones display a speed component in their semantics: soar ('rise quickly'), plunge ('move or fall suddenly'), tumble ('fall quickly and without control'), sweep ('move quickly and powerfully'), meander ('move slowly and not in a straight line'), weave ('move quickly and change direction often'), and Slovene vijugati ('change direction repeatedly and quickly').13 In tumble, the speed component combines with that of unsteadiness and loss of control.
In FM expressions, however, where no actual motion is involved, the speed component obviously cannot express speed but is mapped onto some path feature associated with speed. In the case of vertical motion this feature appears to be the path's gradient, a connection deriving from the observation that the steeper the incline, the shorter the path connecting two extreme positions and, crucially, the shorter the time of travel. The unsteadiness component observed in tumble is mapped onto the very height of the vertically extended entity, a manifestation of the fact that the taller a vertically extended object, the more likely it will lose balance and fall over. The physical properties of mountains thus make it possible for the noun mountain to collocate with tumble.
(12)	Tall green snow-capped mountains tumble majestically into the clear blue ocean waters that make up the Inlet system. (enTenTen)
With manner-of-motion verbs the speed component can be mapped onto the specifics of the path shape. For example, the verbs weave and wind express motion along the same general complex-shape path - a path that is not straight but has several curves, twists and bends. Nonetheless, a weaving path seems to be visualized as having more changes of direction and sharper bends than a winding path, although the two designations can be used interchangeably. The same can be said of the Slovene verbs vijugati14 and viti se, respectively. Again, a connection with speed can be established: the higher the number of curves in a sinusoid-shaped path, the faster one must travel so as to cover a given distance in a given time.
The speed component could also play a role in FM uses of sweep, although the change of direction (13a) or the curved shape of landscape features (13b)15 is more likely to be metaphorically related to the curve involved in a basic sweeping movement — which, in turn, is a quick and forceful one.
13	The meanings given in brackets are based on dictionary definitions (CALD, COBUILD, OALD, SSKJ).
14	'To quickly, continuously change the direction of travel' (SSKJ).
15	(Of landscape) 'to form a long smooth curve' (OALD).
MME1	LANGUAGE 73
(13)	a) The path sweeps left and is joined by another path coming in from the left. (ukWaC)
b) Beyond lay snowcapped mountains and mist-filled valleys sweeping to infinity. (enTenTen)
Of particular interest is the small set of verbs expressing manner of motion without reference to any specific path shape. Wander represents a borderline case here, since its meaning16 still evokes some kind of irregular shape. The verbs hop and slip describe a quick, sudden, effortless movement, the verb ease17 a slow, gentle, careful motion, flow a slow, steady motion, roll a smooth, effortless motion. Meaning components like the effortlessness and smoothness of motion are mappable onto features pertaining to terrain configuration. The connection is obvious: the less difficult the terrain, the smoother/easier motion over it.
For example:
(14)	a) The coast path hops through the car park to follow the lip of Hazelphron Cliff, which terminates at a substantial, wind-swept headland some 600 yards further on. (ukWaC)
b)	The grassy slope slipped away a bit easier on the other side, and the running figure was easily visible. (enTenTen)
c)	A second sloping off, and the path eases its way up this pleasant grass slope with a glorious view of Loch Cluanie. (ukWaC)
d)	The Green Bridge now spans the road and as you walk across it, you would hardly know you were on a bridge at all as the park path rolls across it without interruption. (ukWaC)
To what an extent this type of mapping is actually involved is open to interpretation. It can be supported by an additional element in the sentence structure such as a bit easier in (14b) or without interruption in (14d). Nevertheless, an interplay between the effort and speed components is also possible (cf. (14a), where hop suggests that the path leaves the car park quickly, meaning that the segment running through the car park must be relatively short), or there may be a metaphor underlying the expression. For example, the uses of flow are metaphorically related to the movement of bodies of water like streams and rivers (15a):18
(15)	a) Finally, we reached the ridge, and there, in front of us, was Tuscany, secret valleys flowing from the peaks. (ukWaC)
b) Even an hour away from landing, I'd look out the window and see nothing but mountains rolling one after the other like a stormy ocean of rock. (enTenTen)
In (15b), the landscape configuration is related to a curved shape associated with the basic meaning of roll ('turn over and over and move in a particular direction'): an imaginary sinusoid which is travelled by a(ny) point on the circumference of a circular object rolling along a straight line. It is worth noting that the same type of curve underlies the meaning of undulate, where the connection between the manner of motion and the path shape is fairly straightforward.
16	'To walk around slowly in a relaxed way or without any clear purpose or direction' (CALD).
17	Classified as a motion verb on the basis of uses like I eased my way towards the door (COBUILD).
18	This applies also to actual motion, e.g. Large numbers of refugees continue to flow from the troubled region into the no-man's ¡and (COBUILD).
74 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
E L O P
For example, a swimmer using the undulating3 technique is seen as moving along a sinusoid-shaped path aligned with his/her own body at any given moment. Similarly, an undulating road is automatically visualized in terms of sag and crest curves. Roll, by contrast, does not encode any specific path shape (cf. (I4d) above) — which also explains the difference in classification.
In Slovene, the specifics of the path need to be expressed verb-externally. An exception in this respect is a zigzag path; however, the majority of FM expressions use the adverb cikcakasto 'in a zigzag pattern' (16b) instead of the verb cikcakati 'zigzag' (16a):
(16)	a) Steza spet postane bolj strma in začne cikcakati preko rušnatega pobočja. (slTenTen)
'The path becomes steeper again and starts zigzagging across the grassy slope.' b) Pot se cikcakasto in strmo spušča navzdol in res gre zelo hitro. (FidaPLUS v2) 'The zigzag path descends steeply and it really goes pretty fast.'
Examples (17a—b) illustrate the FM uses of snake and undulate. Slovene employs other means to express the respective meanings. In (17c), the basic-shape manner-of-motion verb viti se 'wind' is combined with the adverb kačasto 'like a snake'. In (17d), the upward/downward motion verbs vzpenjati/spuščati ('ascend/descend') are combined with the adverbs strmo 'steeply' and nenehno 'continuously'.
(17)	a) The road snakes through the hills, bend after bend, and it's time to try "race" mode. (ukWaC)
b)	The road is undulating as it crosses various small hills and ridges left behind by the glaciers. (ukWaC)
c)	Cesta se kačasto vije do visoko ležečih vasi. (FidaPLUS v2)
'The road, like a snake, winds its way up to the villages nestling in the hills.'
d)	Cesta se nenehno strmo vzpenja in strmo spušča. (FidaPLUS v2) 'The road keeps ascending steeply and descending steeply.'
The most frequent constituent conveying path-related information is an adverbial adjunct in the form of an adverbial or prepositional phrase. Another elegant possibility is an adjectival modifier. Sentence (18) below employs both: the adjective kačast 'having the shape of a snake' is fortuitously complemented by the cognate expression kot piton 'like a python' in adjunct function:
(18)	Kačasta cesta se vije po pobočju kot piton. (FidaPLUS v2) 'The serpentine path winds its way up the slope like a python.'
In a similar manner, adjuncts and modifiers in Slovene FM constructions substitute for the meaning components of English verbs that pertain to actual motion, but are mapped onto specific features of the path. A case in point is steepness, which is implicit in the meanings of vertical motion verbs like plunge or soar. Slovene FM constructions use verbs with less specific meanings (cf. rise, fall) and combine them with adverbial or adjectival descriptions pertaining to the gradient.
19 Recognized as a technical term in swimming.
MME1	LANGUAGE 75
(19)	a) Na vzhodni strani se gora precej strmo spušča v dolino. (slTenTen)
'The east side of the mountain descends quite steeply into the valley.'
b)	Od jezera se na tri strani dvigajo strme gore. (FidaPLUS v2) 'From the lake, steep mountains rise in three directions.'
c)	Njena pobočja se dvigajo skoraj navpično. (FidaPLUS v2) 'Its slopes rise almost vertically.'
d)	Njuno vzhodno pobočje pada globoko v dolino Možnice. (FidaPLUS v2) 'Their eastern slope falls deep into the Možnica valley.'
e)	Proti zahodu se pobočje Raduhe spusti tako rekoč prepadno. (slTenTen) 'Towards the west the Raduha slope falls away almost precipitously.'
f)	Steza skozi gozd se hitro spušča. (slTenTen) 'The path through the forest descends quickly.'
(19f) differs from the rest of the examples in that the gradient is expressed indirectly as the result of speed-to-path mapping: the high-speed component of the adverb hitro ('quickly') is mapped onto the steepness of the terrain. A parallel can be found in (20a) below, with the low-speed component of počasi ('slowly') mapped onto the gentle curves of the road. The role of počasi ('slowly') in the description of the road shape is comparable to that of blago ('gently') in (20b):
(20)	a) Široka cesta počasi vijuga proti Veliki dolini. (FidaPLUS v2)
'The wide road winds slowly in the direction of Velika dolina.'
b) Cesta je blago vijugala med grički. (FidaPLUS v2) 'The road wound gently among the hillocks.'
It follows from the discussion that Slovene FM constructions rely heavily on verb-external means to describe the properties of the path. Nonetheless, adjuncts and modifiers are employed by English constructions as well; in fact, they are also found together with verbs whose meanings are specific enough to enable metonymic mapping. In (21) below, for example, the adjunct further specifies the gradient, a property already implicit in the meaning of the verb:
(21)	In the south the mountains plunge vertically into the sea. (ukWaC)
The sentence offers some possibilities for further research, focusing on the role of adjuncts and modifiers in the two languages. Differences can be expected since a verb with metonymic potential makes redundant an adjunct/modifier whose meaning coincides with the meaning expressed by the verb itself (compare plunge ?steeply and plunge vertically). Another related and undoubtedly appealing topic is the metonymic use of adjuncts/modifiers whose manner-of-motion components are mappable onto path properties. Next, the aspectual variants of the Slovene verbs call for a contrastive study on the role of aspect in fictive motion. Last but not least, it follows from the discussion that FM expressions pose several challenges to translation. Rojo and Valenzuela (2003) report informational differences between translating real motion and fictive motion. Stošic and Sarda (2009) question the translational equivalence between fictive motion and posture verbs,
76 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
E L O P
arguing that translating fictive motion changes the conceptualization of the scene. They express the need for more contrastive studies which would help "define cross-linguistically the place of fictive motion in the expression of static scenes" (Stosic and Sarda 2009, 52).
7 Conclusion
English and Slovene differ in the scope of motion verbs conveying path-related information in FM expressions. Both employ directional and manner-of-motion verbs, but their number is considerably smaller in Slovene. The inventory of Slovene verbs includes lexical aspectual variants of directional verbs but lacks equivalents of many English verbs, including those resulting from noun-to-verb conversion. The meanings of English verbs are more specific, displaying components that can be mapped onto certain properties of the path. Besides the obvious correlations in terms of the direction/manner of motion and the path orientation/shape, the following metonymic mappings have been identified, with possible interplay between them: speed —> gradient, speed —> shape, unsteadiness —> height, effortlessness —> terrain configuration. In Slovene FM expressions, these properties need to be expressed verb-externally. The findings call for a study addressing such contrastive differences from the point of view of translation strategies that can be adopted when rendering fictive motion from English to Slovene, and vice versa.
References
Barcelona, Antonio. 2011. "Reviewing the Properties and Prototype Structure of Metonymy." In Defining Metonymy in Cognitive Linguistics: Towards a Consensus View, edited by Réka Benczes, Antonio Barcelona and Francisco José Ruiz de Mendoza Ibáñez, 7—58. Amsterdam: John Benjamins.
Egorova, Ekaterina, Thora Tenbrink, and Ross S. Purves. 2018. "Fictive Motion in the Context of
Mountaineering." Spatial Cognition & Computation, 1-26. https://doi.org/10.1080/13875868.2018.l43 1646.
Fauconnier, Gilles. 1994. Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge, UK: Cambridge University Press.
—. 1997. Mappings in Thought and Language. Cambridge, UK: Cambridge University Press.
Fauconnier, Gilles, and Mark Turner. 2002. The Way We Think: Conceptual Blending and the Mind's Hidden Complexities. New York: Ingram Publisher Services US.
Jackendoff, Ray. 1994. Patterns in the Mind. Language and Human Nature. New York: Basic Books.
Jiménez Martínez-Losa, Noelia. 2007. "Towards a Typology of Fictive Motion Events: Review of Existing Proposals and Presentation of New Perspectives." Interlinguistica 17: 562-69.
Lakoff, George. 1987. Women, Fire, and Dangerous Things. Chicago and London: The University of Chicago Press.
Lakoff, George, and Mark Johnson. 2003. Metaphors We Live By. 2nd ed. London: University of Chicago Press.
Langacker, Ronald W. 1987. Foundations of Cognitive Grammar, Vol. 1: Theoretical Prerequisites. Stanford: Stanford University Press.
—. 2000. Grammar and Conceptualization. Berlin and New York: Mouton de Gruyter.
Matlock, Teenie. 2004a. "Fictive Motion as Cognitive Simulation." Memory & Cognition 32 (8): 1389-1400.
—. 2004b. "The Conceptual Motivation of Fictive Motion." In Studies in Linguistic Motivation, edited by Günter Radden and René Dirven, 221-48. Berlin: Mouton de Gruyter.
MME1	LANGUAGE 77
Matsumoto, Yo. 1996. "Subjective Motion and English and Japanese Verbs." Cognitive Linguistics 7 (2):
Panther, Klaus-Uwe, and Linda L. Thornburg. 2005. "The Role of Conceptual Metonymy in Meaning
Construction." In Cognitive Linguistics: Internal Dynamics and Interdisciplinary Interaction, edited by F. J. Ruiz de Mendoza Ibáñez and S. Peña Cervel, 353-86. Berlin and New York: Mouton de Gruyter.
Rojo, Ana, and Javier Valenzuela. 2003. "Fictive Motion in English and Spanish." International Journal of English Studies 3 (2): 123-50.
—. 2010. "Fictive Motion in Spanish: Travellable, Non-Travellable and Path-Related Manner Information." In
Trends in Cognitive Linguistics: Theoretical and Applied Models, edited by Javier Valenzuela, Ana Rojo and Cristina Soriano, 244-60. Frankfurt am Main: Peter Lang.
Ruppenhofer, Josef. 2006. "Fictive Motion: Construction or Construal?" BLS 32 (1): 307-19. https://doi. org/10.3765/bls.v32i1.3470.
Stosic, Dejan, and Laure Sarda. 2009. "The Many Ways To Be Located in French and Serbian: the Role of Fictive Motion in the Expression of Static Location." In Space and Time in Language and Literature, edited by Marija Brala Vukanovic and Lovorka Gruic Grmusa, 39-60. Newcastle upon Tyne: Cambridge Scholars Publishing.
Takemoto, Masashi. 2010. "Manner-of-Motion Verbs and Subjectification." Journal of Cross-Cultural Studies 4:
Talmy, Leonard. 2000. Toward a Cognitive Semantics. Vol. 1. Cambridge, Mass: MIT Press.
Walinski, Jacek Tadeusz. 2015. "Instrumentality of Fictive Motion in Coextension Paths." Kwartalnik Neofilologiczny 62 (1): 87-101.
BNC (British National Corpus) <https://the.sketchengine.co.uk/> ukWaC (British English Corpus) <https://the.sketchengine.co.uk/> enTenTen (English Web Corpus) <https://the.sketchengine.co.uk/> FidaPLUS v2 (Slovenian Reference Corpus) <https://the.sketchengine.co.uk/> slTenTen (Slovenian Web Corpus) <https://the.sketchengine.co.uk/>
Cambridge Dictionary (the Cambridge Advanced Learner's Dictionary (CALD)) <https://dictionary. cambridge.org/>
Collins Dictionary (the COBUILD Advanced English Dictionary) <https://www.collinsdictionary.com/>
Oxford Learner's Dictionaries (the Oxford Advanced Learner's Dictionary (OALD)) <https://www. oxfordlearnersdictionaries.com/>
SSKJ. Slovar slovenskega knjižnega jezika <https://fran.si/130/sskj-slovar-slovenskega-knjiznega-jezika >
183-226.
15-26.
Dictionaries
78 Frančiška Lipovšek The Role of Motion Verbs in Conveying Path-Related Information in English and Slovene...
m • • ••
•7*< •.'ft
■K' • • •
•V

Begona Lasa Alvarez
University of A Coruna, Spain
2018, Vol. 15 (2), 81-94(120) revije.ff.uni-lj.si/elope doi: 10.4312/elope.15.2.81-94 UDC: 929-055.2(41)"17":81'374
E L
O p
Mary Hays, an Eighteenth-Century Woman Lexicographer at the Service of "the Female
World"
ABSTRACT
The English reformist writer Mary Hays published a compilation of women's biographies entitled Female Biography (1803), with the aim at providing other women with examples to emulate. She intended not only to convey her deepest convictions about women's capacities and abilities, but also to leave her own stylistic imprint on the text. This study seeks to analyse diverse entries of Hays's collection (Lady Dudleya North, Lady Damaris Masham, Margaret Roper, Aphra Behn, and Lady Rachel Russel) in order to elucidate her concerns as a data collector and biographer, and her techniques as a lexicographer, which are chiefly shaped by her concern about education and by her intended audience: women.
Keywords: Mary Hays; Female Biography; lexicography; women's lives
Mary Hays, leksikografinja iz osemnajstega stoletja v službi »ženskega sveta«
POVZETEK
Angleška reformistka Mary Hays je l. 1803 objavila zbirko biografij žensk z naslovom Female Biography, s katero je želela drugim ženskam predstaviti zglede, ki naj jim v svojem življenju sledijo. Poleg Haysinih trdnih prepričanj o sposobnostih in zmožnostih žensk besedilo razkriva tudi njene slogovne posebnosti. Pričujoča študija analizira posamezne vnose v Haysini zbirki (Lady Dudleya North, Lady Damaris Masham, Margaret Roper, Aphra Behn in Lady Rachel Russel) z namenom razjasnitve njenih vodil pri zbiranju biografskih podatkov in njenih leksikografskih tehnik, na katere sta zlasti vplivala njeno prepričanje o pomenu izobrazbe in njeno ciljno občinstvo: ženske.
Ključne besede: Mary Hays; Female Biography; leksikografija; življenja žensk

LITERATURE 81
Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
1	Introduction
In the Monthly Magazine of December 1798, a brief note announced that the English writer Mary Hays (1759—1843) had embarked on a "biographical work of great and lasting interest to the female world, to contain the lives of illustrious women, of all ages and nations" (qtd. in Luria Walker 2013, xii), which was finally published in six volumes in 1803, under the title of Female Biography; or, Memoirs of Illustrious and Celebrated Women of All Ages and Countries: Alphabetically Arranged. This demonstrates the extraordinary work undertaken by the author over at least five years, in which she not only consulted a great quantity of varied sources, but also endeavoured to leave her own imprint on the text she was creating. As the full title makes clear, Hays tried to include in her biographical dictionary celebrated women of disparate backgrounds and occupations. In the preface, she stated that she published accounts of the lives of illustrious women in order to provide her own sex with examples to be emulated (Hays 2013, 3). As a reformist writer, by means of these biographies Hays intended to convey her deepest convictions about women's capacities and abilities and to give visibility to women who had participated in the intellectual, cultural, and political realms of their time (Luria Walker 2006, 215—16).
Hays's opening words in Female Biography offer some meaningful testimonies on the issue of style. First, she quotes Peter Bayle's Dictionnaire historique et critique (1697) to declare that "to abridge with judgment, is of literary labours one of the most difficult" (2013, 8). Additionally, she admits a disdain for "mere compilation [...] aiming at a clear, correct, and even harmonious style" (2013, 8). As her "book is intended for women, and not for scholars," her purpose is "to collect and concentrate, in one interesting point of view, those engaging pictures, instructive narrations, and striking circumstances" (2013, 7). Did Mary Hays's biographies achieve these goals? The current study, while not comprehensive, seeks to provide an answer to this question by means of a preliminary approach to Hays's stylistic concerns as a data collector and biographer, and to her techniques as a lexicographer.
2	Hays as a Vindicator of Women's Education
Hays was born in a middle-class family of Rational Dissenters, and the death of her father, among other early misfortunes, was a determining factor in her life, as she gained first-hand of women taking on more responsibility in a patriarchal society. Later on, when she met Mary Wollstonecraft and other members of radical circles, she became acquainted with their revolutionary ideals,1 while she also received literary and emotional support to revive and continue her career as a writer. Hays thus enjoyed the positive influence of a cultural and intellectual milieu, in which women had the opportunity to participate in gatherings and debates and express their ideas. However, she quickly realized that women's role in the world of letters was that of outsiders (Luria Walker 2006, 1).
1 Hays and other writers of the 1790s, including William Godwin, Thomas Holcroft, Elizabeth Inchbald and Charlotte Smith, received the name of Jacobin novelists, as they focused on the positive consequences that the French Revolution could bring to English society. However, their intention was not revolutionary, as they simply called for a government guided by reason, and not by money or rank (Kelly 1976, 8).
82 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
E L O P
Interestingly, Wollstonecraft acted as a mentor for Hays, the influence of the former's A Vindication of the Right of Woman (1792) being especially strong on the latter, who consistently called for greater advancement and autonomy for women. Education was essential in this context, and Hays always demanded better instruction for women in her writings. Wollstonecraft and Hays were in turn admirers of catharine Macaulay, a historian and activist, who shared with numerous women of the period a concern about women's education.2 Macaulay was well-known for her multivolume History of England (1754—1761); however, she also wrote an essay entitled Letters on Education, in 1790, in which she attacked the most relevant and widespread educational systems of that time, namely those proposed by Rousseau and Hume. According to these two men, the education that girls should receive was based on the association of women with lesser virtues, such as delicacy, smoothness, beauty and dependence (Frazer 2011, 608), which made them incapable of receiving the same education as boys. Macaulay, however, was in favour of an egalitarian education, for knowledge, in her opinion, was not determined by gender and thus she claimed that the assumed distinctions between men and women "do not in any manner proceed from sexual causes, but are entirely the effects of situation and education" (1790, 202).
In her collection of essays, Letters and Essays, Moral, and Miscellaneous (1793), Hays similarly expressed her indignation about certain common attitudes towards women, which not only damaged them, but also the whole of society: "By degrading the female character, and by repressing the wish for improvement, men have, with a narrow and temporary policy, been enemies, not only to their own happiness, but to their offspring" (1793, vii). Years later, in an article she published anonymously in 1797, "Improvements suggested in Female Education", Hays describes the terrible habit adopted by many families of her time, which consisted of bringing up their daughters just for "the precarious chance" of marriage (1797, 194). The consequence for these young women was victimization and helplessness. However, educated women would be able to practice diverse trades, including liberal arts, arithmetic and bookkeeping, which would permit them to support themselves and not to depend on men. To achieve this and defeat inequalities, Hays asks for government reforms and new policies, which should focus on individuals' "faculties and talents, without distinction of rank or sex" (1797, 195). Following the same path, Hays's two most well-known novels, Memoirs of Emma Courtney (1796) and The Victim of Prejudice (1799), can be described as compendia of all the injustices committed to the middle-class women of her time (Ty 1998, xvi), while in Female Biography she looked to women in the past in order to offer powerful examples to her readers, mostly women, so that they could fight prejudice and discrimination. Most of these examples were upper-class women, who had a better access to education, and thanks to the opportunity they enjoyed, were able to be and act as great women, and able to impact society (Luria Walker 2005, 255). Furthermore, Hays used some of these women's life stories— e.g. those of Lady Masham and Margaret Roper — to introduce her feminist viewpoints, particularly the issue of women's education, as will be discussed below.
2 In her Vindication of the Rights of Woman, written just after Macaulay's death, Wollstonecraft declared this about her: "The woman of the greatest abilities, undoubtedly, that this country has ever produced. —And yet this woman has been suffered to die without sufficient respect being paid to her memory", and she added that she was "an example of intellectual acquirements supposed to be incompatible with the weakness of her sex" (1999, 180). Hays was one of her first biographers and being herself very conscious of the prejudiced viewpoint with which some women writers were regarded and judged, stated about Macaulay: "A female historian, by its singularity, could not fail to excite attention: she seemed to have stepped out of the province of her sex; curiosity was sharpened, and malevolence provoked. The author was attacked by petty and personal scurrilities, to which it was believed her sex would render her vulnerable" (2014, V, 294). Indeed, Macaulay's radical political ideas and some personal circumstances affected her consideration and esteem among the public (Hill 1992, 117).
LITERATURE 83
Although women's education had been a controversial issue before, in the eighteenth century it became a particular point of contention, as the literary realm was more and more populated by female writers, and the rate of literacy among women was rising rapidly. Indeed, reading was not only more accessible to women but it also started to be considered a suitable activity for them. As Nancy Armstrong explains, during the eighteenth century a new model of woman emerged, confined to the domestic realm with the only task of supervising the household servants' work. and consequently such women had large amounts of leisure time (1987, 79). Many thus spent their spare time reading, although this new habit did eventually cause great anxiety, since it was more and more difficult to control what women read. Indeed, it was considered that women, due to their allegedly weaker disposition and vulnerability, were particularly susceptible to some materials and topics. Nevertheless, history was recommended as very beneficial for women's instruction, and among historical writings biographies of famous women were particularly popular. Moreover, intellectuals and scholars proposed this type of texts instead of fiction, since they gathered real examples and models for women readers to imitate (Pearson 1999, 51).
3 Hays as a Data Collector and Lexicographer
After the publication of Female Biography, both the structure and pattern of Hays's text were considered problematic, and thus a target of critical remarks by reviewers. Her selection of women subjects and the alphabetical arrangement of entries introduced a transgressive turn into the text, in that worthy and reprehensible women were set together. While this is a notable feature of the work, it was roundly criticised due to the negative effects that it might have on its intended readership (Luria Walker 2005, 330-33; Wood 1998, 117-18). The eighteenth century saw a huge number of compendia of this sort, and information in them was frequently presented in a logical, accessible manner. Indeed, most of Hays's contemporaries "favoured works in alphabetical order designed for rapid reference, rather than didactic works arranged thematically. Erudition was being displaced by efficient information storage and retrieval" (Headrick 2000, 172), so in adopting this arrangement she shows herself to be in sync with her times.
Similarly, more recent studies have expressed reservations as to the form of this compilation, considering it hack work, undertaken for mere financial gain (Kelly 1993, 246). Interestingly, Samuel Johnson's entry for "Grub Street" in A Dictionary of the English Language (1755) states that this well-known street in London was inhabited by literary hacks, mediocre and needy writers "of small histories, dictionaries, and temporary poems" (Encyclopaedia Britannica 2018), and includes dictionary compilers among the category of second-rate writers who worked chiefly for money. Taking into account the interests of both Hays and her publisher, the financial motivation for this work is indeed beyond doubt (Luria Walker 2006, 216; Luria Walker 2013, xii). Moreover, Hays's approach in the composition of her biographical dictionary took an inevitable course, and could hardly have been differently. Certainly, Isabel Rivers in an analysis of various eighteenth-century biographical dictionaries has explained the procedure for writers of such texts, which was based on borrowing; they "drew on earlier ones, or [...] took account of other dictionaries published in the meantime which commented on the earlier version. There would seem to be no end to the task of the modern historian hoping to disentangle the chain of borrowing, allusion and criticism" (2003, 137). This aspect is particularly evident in the case of Hays, who includes her sources at the end of most entries. And despite not citing every source, the enlightened spirit and empirical method in her compiling work is palpable (Capern 2005, 152). Moreover, at that time any borrowing could be justified in terms of a successful stylistic improvement, and indeed textual unity and seamlessness were the main features when
84 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
E L O P
evaluating literary works (Mazzeo 2007, 3—4). In this regard, Hays's text exemplifies such aims as refinement, together with the author's own opinions on her personal style, particularly regarding her target audience, women.
Susan Staves (2007) has raised the question of how eighteenth-century women writers got access to all the books they read, and considering the number of sources quoted by Hays, this question certainly arises in this case. Being a woman, and one of limited means, the option of purchasing so many books was beyond her. Moreover, women were excluded from almost all of those significant places where books could be read and consulted, including schools, universities and libraries, particularly if they were also dissenters, as with Hays. Nevertheless, the new paradigm established in the literary system of the time offered women new opportunities, and they were able to participate in this system as givers and receivers of books with "friends and colleagues engaged in a common project" (Staves 2007, 208). As an active member of various circles of like-minded writers and friends, Hays might have access to the private libraries of some of these (Spongberg 2014, 539),3 and particularly of those who were or had been engaged in similar projects. Staves mentions a further way in which women might have obtained books, one which again might have been exploited by Hays: some writers obtained books as a form of patronage by publishers, in that a supply of such books constituted the sort of material writers needed for the work they were doing for those very publishers (Staves 2007, 196). Thus Richard Philips, the publisher of Female Biography, might have supplied Hays with some of the biographical texts he had already printed; as a businessman interested in the increasingly popular genre of life-writing, this served his commercial purposes well (Luria Walker 2006, 216).
The title of the text suggests that Hays's primary purpose was very broad; although her own country remains a key element, given that approximately one third of the women recorded are British, and some of these women's biographies are among the most extensive ones. In terms of the form of the biographies, there is no uniform pattern, and the length of the entries varies considerably, from just a few sentences to essays of more than 100 pages, the latter being the case for both Elizabeth I and Mary Queen of Scots. However, most of the British women in Hays's text are not royal, and the most common length of the entries for these is far shorter. Some 94 British women are described (of the total of 300 entries for the whole work), most of these being noble and learned ladies, and a considerable number of them also writers (Lasa Alvarez 2014). There is one explanation of this: the sources used by Hays for the great majority of these entries. Clearly, if she wanted to produce a compilation of women's lives, her best approach would be to draw on existing texts of similar nature, which included principally "female worthies," that is, the kind of historical work which embraced those various female monarchs, warriors, saints, and savants distinguished by their rank, talent, piety, or otherwise noteworthy lives (Guest 2000, 49-69; Chernock 2013, 115-36; Hicks 2014, 18-33).
First, the crucial role of George Ballard's Memoirs of Several Ladies of Great Britain (1752) has to be recognized. This text is not only one of the most repeated sources in Hays's entries, but also the one which marked the emergence of a canon of worthy women, and "the key source for information about women writers in the Renaissance and seventeenth century for almost all subsequent biographical dictionaries including women writers, [... and] for biographical notices in anthologies of women writers" (Guest 2000, 50, 78). Ballard was captivated by the
3 Hays had already enjoyed access to the private libraries of friends, such as Mrs Collier, when she was younger (Luria Walker 2006, 12).
LITERATURE 85
antiquarian method of recovering historical data and materials, but usually did not go on to make any detailed of analysis of the facts established (Chernock 2000, 119—20). As an antiquarian, his work lay mainly in collecting and gathering as much information as possible for each entry, including sources such as archival records, letters, wills, sermons, elegies, manuscripts, and even inscriptions on tombstones and monuments (Sweet 2004, xvi)4.
The anonymous Biographium Faemineum: The Female Worthies (1766) is also a source text for Hays. It reproduces many entries from Ballard's Memoirs, but in having a more inclusive objective, as set out in its title and preface, the collection includes foreign women and also some historical protagonists of scant virtue (1766, viii), such as Cleopatra and Aphra Behn.5 Finally, Thomas Gibbons's Memoirs of Eminently Pious Women (1777) should be also noted here. For Gibbons, a dissenting minister, religion and virtue were the salient criteria for his choice of subjects (1777, vii).
Evidently the use of these biographical materials, which incorporated the authors' didactic purpose and ideological point of view, stressing such features as women's virtue and modesty, affected and sometimes shaped the way some of these women have subsequently been regarded. The women included in these authors' collections were celebrated for diverse reasons; however, their portrayals consistently included some sort of self-denying accomplishment, which was a prerogative of women's private roles as mothers, daughters and wives. They perhaps also influenced Mary Hays herself; it is thought that as a consequence of her use of earlier sources of this kind, she moved away from her previous radical and liberal views, to adopt a more conservative perspective and tone in her biographical compendium, more in accordance with the post-revolutionary atmosphere in Britain at the time (Kelly 1993, 234; Spongberg 2002, 117).
4 Hays as a Biographer: Some Examples
The biography of the Honourable Lady Dudleya North (1675—1712) is a good example of Hays's system of textual assembly and work. Lady Dudleya was member of a significant noble family actively involved in government and politics during the troubled second half of the seventeenth century in England. Despite this, she lived a mainly secluded life, devoting her entire life to study. However, thanks to her family's high position in society she was able to collect a large number of books, a pursuit which she shared with some of her relatives, such as Sir Roger North, one of her educated and cultivated uncles, a lawyer and biographer, and a member of the King's Counsel. The entry is just one page long, and at the end of it Hays offers a brief list of sources, which include "Ballard's British Ladies" and "Biographium Faemineum" (2014, 527). Lady Dudleya North's biography is reproduced with minor changes from Ballard's version, which is clearly also the source of the other text mentioned by Hays, Biographium Faemineum.
The process of summarising is clear and progressive, concluding with Hays's version, which, apart from being shorter, is, in my opinion, clearer, simpler and smoother than Ballard's original. Hays's biography is also reordered, following a more logical scheme. Thus Hays takes the section which explains that Lady Dudleya owned a notable collection of books and that it was donated to the parochial library after her death, and places this at the end of the entry, as these events occurred after the subject had died. It is, then, a simple but effective improvement in the presentation of information.
4	This collection also includes numerous exemplary women writers (Lasa Alvarez 2015).
5	The inclusion of such women in biographical collections was frequently accompanied by warning labels by the compilers (Booth 2004, 67), as Hays herself did in presenting Aphra Behn's life.
86 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
E L O P
In the case of a longer entry, such as Lady Damaris Masham (1658—1708), a philosopher of the late seventeenth and early eighteenth centuries, the same sources are used. Again, Hays and the anonymous author of the Biographium Faemineum both make use of Ballard's compilation. The difference in style is also clear in this case, since the anonymous compiler generally maintains the sentence structure of Ballard's text, while Hays tries to build a more flowing style. It is interesting to note in more detail the dissimilarities in the approaches of these two authors with respect to Ballard's text at the point where the latter reproduces Lady Masham's own words on education. Indeed, almost half of Ballard's entry concerns allusions to and quotations from one of her treatises: Occasional Thoughts in reference to a Virtuous Christian Life (1705), and particularly those aspects related to the education of both women and men. However, in the Biographium Faemineum the space assigned to it is no more than a short reference: "The principal design of which (education) is, to improve religion and virtue; and need it is so full of instruction, that, if carefully perused by both sexes, it could not fail of obtaining much of its desired end" (1766, 126), which might be seen as an invitation to read the book.
Mary Hays also summarises Ballard's text, but contrary to the author of Biographium Faemineum, and due to her personal interest in this matter, as previously examined, she draws on Masham's opinions to offer her readership another perspective on the necessity of women's education. As a friend and follower of Locke, Masham placed great importance on the civic and moral role of education in society, yet she focused her interest on women's education, and particularly on the role of cultivated mothers as educators of worthy citizens for an enlightened society (Hutton 2010, 74).6 After giving a general idea of Masham's opinion on the "neglected education of daughters, among persons of high rank, and the frivolity and dissipation in which they consume their lives," from the excerpts selected by Ballard, Hays chooses those parts in which gentlemen's mistaken ideas about education induce them to ridicule "a learned lady." This is particularly reprehensible, she argues, in that a learned lady is "capable of instructing her children, and of implanting in their minds just principles" (2014, 496-97).
There are several more entries in Hays's compilation based on the same sources, which illustrate the extent of her reliance on these two texts. Yet also very clear is her desire to infuse her text with her own style, and to emphasise those assumptions which she considers instrumental for the general aim of her biographical dictionary. Her personal interest here is seen in occasional additions to her core sources. A case in point is the biography of Margaret Roper (1505-1544), the cultivated daughter of Thomas More, Lord High Chancellor during the reign of Henry VIII. At the end of the entry Hays notes her usual sources, Ballard and Biographium Faemineum; however, the initial long paragraph of the entry is her own:
In favour of the liberal cultivation of the minds of women, it may be observed, that at no period of the English history does there appear to have been greater attention paid to the culture of the female mind, than during the age of Elizabeth; and at no time has there existed a greater number of amiable and respectable women. [...] Among women distinguished for their virtues and acquirements, in the 16th century, the three daughters of Sir Thomas More hold an elevated rank (2014, 92-94).
On the following page, she adds an exclamation contrasting those favourable times for women's education to her less cultivated present-day gatherings: "What a charming picture, contrasted
6 When the notion of education was associated with women, this tended to imply domestic education, a task often undertaken by mothers, who by the end of the eighteenth century had acquired the role of the guardians of national knowledge (Davies 2014, 6-7). This issue was also emphasized by Mary Wollstonecraft (Kelly 1993, 243).
LITERATURE 
87
with modern seminaries ofvanity and dissipation!" (2014, 94). Curiously enough, Hays considers More's daughters as belonging to "the age of Elizabeth," while in Roper's biography several references to King Henry VIII, the reigning monarch at that time, can be found, particularly regarding the refusal of Roper's father to swear the oath of Supremacy and its fatal consequences. It is quite clear from her words in the entry that Hays was not overly fond of Henry VIII, describing him as "a capricious tyrant" (2014, 98).7 As in the entry for Elizabeth I, various instances of the Queen's cultivated court and her inclination for study as part of the customs of the times are noted in Margaret Roper's entry, despite belonging to a previous historical moment.
In addition to using compilations of women's biographies, such as those mentioned above, Hays also drew on generalist collective biographies of the period (Spongberg 2014, 539), as her biography of the writer Aphra Behn (1640—1689) shows. This entry is somewhat contradictory, in that Hays initially calls her the English Sappho (2013, 385), and whereas she does recognize Behn's literary merits, her works are said not to be recommended, a somewhat perplexing and incongruent position in dealing with a writer. Indeed, Behn's works are "reprehensible for their licentiousness" and "do not serve the cause of virtue" (2013, 384—85). Hence, the sources assembled by Hays to construct Behn's portrait are of key interest here. At the end of her entry, just two are cited, "Biographia Britannica" and "Biographium Faemineum" (2013, 386). The former might refer to the biographical dictionary in six volumes, edited by William Oldys and published for the first time in England between 1747 and 1766. However, given the personal acquaintance between Mary Hays and Andrew Kippis, with both writers frequenting the same Dissenting circles (Spongberg 2010, 254), it is more than logical to think that she used the second edition of this compendium, which had been edited by Kippis.
A comparison of entries for Behn in both editions with Hays's own shows that she did indeed draw on the second edition, as we will see below. This new edition bears the same title, with the addition of the following information on the title-page: "The Second Edition, with Corrections, Enlargements, and the Addition of New Lives: By Andrew Kippis, D.D.F.R.S. and S.A. With the Assistance of the Rev. Joseph Towers, LL.D. and other Gentlemen" (1780, n.p.). It is in fact an incomplete edition of five volumes which appeared between 1778 and 1793. As already noted, Hays based her biographical dictionary on the guiding principles established by Bayle's in his influential work, as the editors of the Biographia Britannica also did. Indeed, Hays's close familiarity with Bayle is beyond doubt; Luria Walker pointed out that she was introduced to Bayle by Robert Robinson, another Dissenting scholar and minister (2013, xxiv). The second source for Behn's entry is Biographium Faemineum, as for other entries already discussed here.
An analysis of Hays's entry on Aphra Behn makes it clear that she extracted most of the data from the Biographia Britannica. Interestingly, the actual biography of Behn in this text is quite short, yet the scholarly footnotes with additional information and explanations occupy much more space (1780, 141—46).8 While the anonymous author of the Biographium Faeminum reproduced almost word for word the main text of the Biographia Britannica, Hays took both its components and rearranged them in her entry, thus offering a much more readable text. As a consequence of the use of this source, which combined biographical information and fictional excerpts, in Hays's text some material taken from Behn's well-known narrative Oroonoko; or, The Royal Slave (1686) is used as if it is an authentic and true account. Hays goes as far as mentioning a possible sexual relationship
7	An expression that Hays had used before in her biography of Queen Elizabeth I (2014, 73).
8	The editors of this biographical dictionary employed the same method used by Bayle in his text: the main text for the historical facts, and footnotes for discussion, commentaries and additional curious information (Rivers 2003, 138—39).
88 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
E L O P
between Behn and this prince. As happened in most of the early biographies of Behn, this affair is then refuted in Hays's text and described as a "calumny" (2013, 383).9 However, as Jane Spencer has explained, this denial "serves, of course, to draw attention to the titillating idea of a sexual relationship" (2000, 35). Similarly, Hays mentions "Some letters, addressed to a gentleman, whom she poetically calls Lycidas; are printed in her Memoirs: this attachment appears to have been not less tender and ardent than unfortunate" (2013, 383). The letters are interpreted biographically, when it is not entirely certain that they are even hers, yet again they contribute to introduce the reader to Behn as a passionate and desperate woman in love (Spencer 2000, 38).
Generally speaking, Behn's critical reception in the eighteenth century was negative, usually appearing in contrast to other women writers, such as Katherine Philips or Elizabeth Singer Rowe, who epitomised chastity and piousness (Todd 1998, 176). Hays, like her sources, adheres to this tendency and reiterates conservative comments about women's role in the public sphere (Wood 1998, 133-34). Nevertheless, it has also been noted that the mere inclusion of Behn in the compilation, together with the relative length of the entry, is significant and remarkable, given that this writer was conspicuously absent from most biographical collections of the period, more concerned as they were with presenting to their intended readers depictions of virtuous and pious women (Wood 1998, 132-33). Also, bearing in mind that Hays used the second edition of the Biographia Britannica as her main source to reconstruct Behn's life, her comments, while sometimes quite unfavourable, are much more benevolent and considerate than those expressed by one of the editors of the second edition of the Biographia Britannica, who used the final footnote to add new information about Behn, which is full of critical remarks on her works and character. Indeed, this latter editor, identified as K (perhaps Kippis himself),10 seems to congratulate himself on the fact that Behn's writings are "now little regarded, her novels excepted, which, we suppose, have still many readers among the unhappily too numerous a class of people who devour the trash of the circulating libraries" (1780, 146). Finally, he responds to a certain gentleman on the appropriateness of including in the Biographia Britannica such a character as Behn, saying that she had to be part of it, in that she was undeniably a celebrity in her times and her works very famous, "though, in her indelible disgrace, her talents were prostituted to licentious purposes" (1780, 146). Hays attempts to give a more tempered view of Behn's life and mentions "the manners of the times [...] and the contagion of the court" (2013, 385) as the causes for the censure of Behn's works, giving voice to the eighteenth-century construction of the writing in the Restoration period as decadent and salacious (Spencer 2000, 2). However, also, as Kelly has observed, Behn's biography might also serve to illustrate what is perhaps the most significant issue which Hays condemns in this compendium, "the historic corruption and trivialization of women in court culture" (1993, 239).
Finally, Hays also uses specific sources for individual entries, as is the case with Lady Rachel Russel (1636-1723), a rather long entry of 15 and a half pages, this reflecting her active and public life, and her longevity. The sources assembled by Hays here are "Hume's History of England", "Letters of Rachel, Lady Russell, from the MS in the Library at Wooburn Abbey" and "Memoirs of Pious Women" (2014, 346). That is, a recently published work of general history: David Hume's History of England (1754-1761), a specific text related exclusively to the subject of the entry: Letters of Lady Rachel Russell; from the Manuscript in the Library at Woburn Abbey (1773),
9	Hays takes the excerpt of Oroonoko from the corresponding footnote in Biographia Britannica, and she subsequently repeats the same procedure in the rest of the entry.
10	In the preface to the second volume of the Biographia Britannica, in which Behn's biography is included, some of the original writers are identified by an initial letter, but nothing is said about this editor whose signature is K (1780, viii).
LITERATURE 89
and the aforementioned Memoirs by Gibbons. Hays's approach to the use and organization of these sources is significantly different here. The main source is one text, which specifically deals with the woman in question, or better said, with her husband Lord William Russell. The text is an anonymous and favourable of his character, which was published as a preface to The Letters of Lady Russell, a collection that appeared in 1773, 50 years after her death. Hays, however, might have been first acquainted with her subject through Gibbons, one of the first authors to include Russell's biography in a compendium of this type. Although Lady Rachel was well known during her lifetime, and indeed was described by a contemporary as "One of the best of women", her life went practically unnoticed until her letters were published with the above-mentioned preface in 1773, after which her life enjoyed immediate popularity and was incorporated in women's biographical collections, as well as in other historical and literary texts, such as plays, and even in pictures (Schwoerer 1988, xxiv-xxvi; Culley 2014, 34-52).
The interest in Lady Russell's biography is a telling instance of what Paula Backscheider has noted about women and biography: "women in order to be worthy subjects of biographies, must marry into, fall into, or fashion a worthwhile role in life" (Backscheider 2001, 144). Lord Russell was Lady Rachel's second husband, a man who participated actively in English politics as a member of the Country Party, the forerunners of the Whigs. He allegedly took part in a conspiracy, the 1683 Rye House Plot, which sought to raise a rebellion by seizing King Charles II's guards, and thereby to kill him. On the 26th of June, Russell was arrested and on the 13th of July tried and found guilty of high treason. He was sentenced to death and executed on the 21st of July 1683.
Lady Russell's letters came to light just after the publication of John Dalrymple's Memoirs of Great Britain and Ireland. From the Dissolution of the Last Parliament of Charles II, until the Sea-battle off La Hogue (1771-3). Dalrymple's work is the first history of late seventeenth-century Britain based on French sources, and it revealed that Lord Russell had plotted with members of the French court. This revelation damaged Russell's reputation and that same year the book including his wife's letters and the anonymous preface was published (Schwoerer 1988, xxiv; Culley 2014, 35). As mentioned above, most of the information in Hays's entry for Lady Russell comes from the Introduction to the Letters, at least in terms of the information on Lord Russell's involvement in the Rye House Plot, his arrest, trial, sentencing and execution; however, most of this information in turn derives from other sources quoted by the anonymous author of this Introduction. These include some of the best known historians and history books of the period: Bishop Burnet's History of His Own Time (1724), Paul de Rapin's L'Histoire d'Anglaterre (1724), and Thomas Birch's The Life of the Most Reverend Dr. John Tillotson (1725), among others. Additionally, on account of the main source employed, Hays's biography of Lady Russell is loaded with the ideology that pervades that text, vindicating Lord Russell and the Whigs and blaming the encroachments of the crown. For instance, Hays argues:
No one doubted the innocence of Russell respecting the charge of conspiring against the life of the king, which he solemnly denied with his dying breath. [...] his principal guilt had been his opposition in parliament to what he deemed unconstitutional measures, with his effort for the exclusion of the duke of York from the throne (2014, 336).
Although devoting almost half of Lady Russell's entry to the circumstances surrounding her husband's trial and execution, Hays takes the opportunity to introduce Lady Rachel within this account, given that she played an active role in her husband's trial and the events that followed. Hays describes how Lady Russell acted as an assistant to her husband, taking notes during the trial, at the express wish of Lord Russell himself, and the dramatic reaction of the assistants,
90 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
E L O P
who "melted in tears"; and later noting that "Lady Russell threw herself at the feet of the king, and pleaded with tears the merits and loyalty of her father, as an atonement" for the possible offences (2014, 335). The description of the couple's final parting is also extremely moving, in which she conducted herself with "heroic fortitude", and was said by her husband to have been a blessing to him (2014, 340). In fact, the attribute of fortitude has been a commonplace in the numerous biographical accounts of Lady Rachel, and it has been explained by her biographers as a combination of maternal and family duties with courage and dignity (Culley 2014, 38).
Lady Rachel survived her husband by 40 years, and thus the events after his death were taken not from the aforementioned Preface, but from her own letters plus annotations and appendixes added to these. Notable here is Lady Rachel's resilience on the death of one of her daughters, which appears in a footnote. Her youngest daughter, Catherine, died in childbirth, while the eldest one was also giving birth, and she had to cope with these tragic circumstances. In this case, Hays does not summarise or foreshorten the passage, with the clear purpose of conveying every dramatic feature of the story. Other significant episodes rendered by Hays include a dangerous incident involving a candle made with gunpowder, which Russell lit by mistake when reading in her bedroom, and information about her cataract operation.
5 Conclusion
The analysis presented above shows that Hays created a completely new biography for Lady Rachel, since Gibbons, (her other biographer), provided a much shorter entry in his collection, with fewer details, though he did incorporate some of her letters, unlike Hays. Whereas in other entries, Hays summarized the information she found in her habitual sources, in Lady Russell's case she had to make a more deliberate selection of exactly what she wanted to include in her entry. Thus this entry might be considered as coming closest to her portrait of an ideal woman: virtuous and pious, an affectionate daughter, wife and mother, who is able to endure life's trials with courage, values which are wholly in agreement with those of women typically portrayed in the "female worthies" kind of compilation, and with the type of historical characters that in Hays's opinion women readers seemed to enjoy the most:
Women, unsophisticated by the pedantry of the schools, read not for dry information, to
load their memories with uninteresting facts, or to make a display of a vain erudition. A
skeleton biography would afford to them but little gratification: they require pleasure to be
mingled with instruction, lively images, the graces on sentiment, and the polish language
(2013, I, iii—iv).
This brief analysis has shown that Mary Hays's labour as a lexicographer in Female Biography is not an example of mere "hack work," and that she tried to simplify and summarise her sources, as well as to select those aspects which might entertain and even shock her readers the most, while also being both educative and improving in nature. It thus seems clear that she always had her female readers in mind, having expressed in the Introduction to the work the assumption that women's "understandings are principally accessible through their affections" (2013, iv). In order to stir her readers' hearts, as has been examined in the article, Hays clearly exploits the most dramatic events in her subjects' lives, while in some other life-accounts, such as Behn's, she prefers to concentrate on salacious love affairs.
Furthermore, she proposed models of women "to excite a worthier emulation" (2013, v) in her audience, with the main purpose of providing instruction and combating prejudice against
LITERATURE 
91
women. Hays and other women of her time, including Mary Wollstonecraft, Catharine Macaulay, Helen Maria Williams, Anna Laetitia Barbauld, Mary Robinson, Charlotte Smith, and Elizabeth Inchbald, were pejoratively called "unsex'd females" by T.J. Mathias and Reverend Polwhele, well-known representatives of conservative thought, due to their radical and revolutionary ideas.11 Hays in particular suffered strong personal criticism, in that she was, for instance, the target of Elizabeth Hamilton's satirical novel, Memoirs of Modern Philosophers (1800). More importantly, although in Female Biography Hays's lexicographic work was predominantly influenced by her sources, she was able to reveal quite often her personal feminist philosophy, which was rooted in the necessity of an egalitarian education for women and in forceful criticism of gender roles in the patriarchal English society of her time.
Acknowledgements
This article has been possible thanks to the support of the Research Group "Literatura e cultura inglesa moderna e contemporánea" (G000274), funded by the University of A Coruña, and the Research Network VINS "(Neo)Victorian Network in Spain" (MINECO, FFI2015-71025-REDT).
References
Anon. 1766. Biographium Faemineum: The Female Worthies, Or, Memoirs of the Most Illustrious Ladies of All Ages and Nations Containing (exclusive of Foreigners) the Lives of Above Fourscore British Ladies. Collected from History, and the Most Approved Biographers, and Brought Down to the Present Time. London: Printed for S. Crowder, et al.
Armstrong, Nancy. 1987. Desire and Domestic Fiction. A Political History of the Novel. New York: Oxford University Press.
Backscheider, Paula R. 2001. Reflections on Biography. Oxford: Oxford University Press.
Ballard, George. 1985 [1752]. Memoirs of Several Ladies of Great Britain, Who Have Been Celebrated for Their Writings or Skill in the Learned Languages, Arts and Sciences, edited by Ruth Perry. Detroit: Wayne State University Press.
Booth, Alison. 2004. How to Make It as a Woman. Collective Biographical History from Victoria to the Present. Chicago: The University of Chicago Press.
Capern, Amanda L. 2005. "Enlightenment." In Companion to Women's Historical Writing, edited by Mary Spongberg, Barbara Caine, and Ann Curthoys, 152—60. Basingstoke: Palgrave.
Chernock, Arianne. 2013. "Gender and the Politics of Exceptionalism in the Writing of British Women's History." In Making Women's Histories: Beyond National Perspectives, edited by Pamela S. Nadell and Kate Haulman, 115—36. New York: New York University Press.
Culley, Amy. 2014. "Reading the Past. Women Writers and the Afterlives of Lady Rachel Russell." In Historical Writing in Britain 1688-1830. Visions of History, edited by Ben Dew and Fiona Price, 34—52. Basingstoke: Palgrave.
11 Polwhele published anonymously in 1798 a poem entitled The unsex'd females: a poem, addressed to the author ofthe Pursuits of literature. He referred to Hays as follows: "flippant HAYS assum'd a cynic leer", and in a footnote he further explained that although this writer is little known, "she is evidently a Wollstonecraftian" (qtd. in Luria Walker 2005, 249), using this term with a derogatory meaning. Polwhele was inspired by T.J. Mathias, to whom he addressed his text. Indeed, Mathias in The Pursuits of Literature (1798) had written about "our unsex'd female writers (who) now instruct, or confuse, us and themselves, in the labyrinth of politics, or turn us wild with Gallic frenzy" (qtd. in Pascoe 2004, 213).
92 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
Davies, Rebecca. 2014. Written Maternal Authority and Eighteenth-Century Education in Britain. Educating by the Book. Farnham: Ashgate.
Encyclopaedia Britannica Online. 2018. "Grub Street. Literary Hacks." https://www.britannica.com/topic/ Grub-Street.
Frazer, Elizabeth. 2011. "Mary Wollstonecraft and Catharine Macaulay on Education." Oxford Review on Education 37/5: 603-17. https://doi.org/10.1080/03054985.2011.625165.
Guest, Harriet. 2000. Small Change. Women, Learning, Patriotism, 1750-1810. Chicago: The University of Chicago Press.
Gibbons, Thomas. 1777. Memoirs of Eminently Pious Women Who Were Ornaments to Their Sex, Blessings to Their Families, and Edifying Examples to the Church and World. London: Buckland.
Hays, Mary. 1793. Letters and Essays, Moral, and Miscellaneous. London: T. Knott.
—. 1797. "Improvements Suggested in Female Education." The Monthly Magazine and British Register 3: 193-95.
—. 2013 [1803]. Female Biography; or, Memoirs of Illustrious and Celebrated Women, of All Ages and Countries. Alphabetically Arranged, edited by Gina Luria Walker. Vols. I—III. London: Pickering & Chatto.
—. 2014 [1803]. Female Biography; or, Memoirs of Illustrious and Celebrated Women, of All Ages and Countries. Alphabetically Arranged, edited by Gina Luria Walker. Vols. IV-VI. London: Pickering & Chatto.
Headrick, Daniel R. 2000. When Information Came to Age. Technologies of Knowledge in the Age of Reason and Revolution, 1700-1850. Oxford: Oxford University Press.
Hicks, Philip. 2014. "Female Worthies and the Genres of Women's History." In Historical Writing in Britain, 1688-1830. Visions of History, edited by Ben Dew and Fiona Price, 18-33. Basingstoke: Palgrave.
Hill, Bridget. 1992. The Republican Virago. The Life and Times of Catharine Macaulay, Historian. Oxford: Clarendon Press.
Hutton, Sarah. 2010. "Masham, Damaris (née Cudworth)." In The Continuum Companion to Locke, edited by S.-J. Savonis-Wroth, Paul Schuurman and Jonathan Walmsley, 72-76. London: Continuum.
Kelly, Gary. 1976. The English Jacobin Novel 1780-1805. Oxford: Clarendon Press.
—. 1993. Women, Writing and the Revolution 1790-1827. Oxford: Oxford University Press.
Lasa Alvarez, Begona. 2014. "The Discourse in Compilations of Women's Biographies in the Eighteenth Century: Constructing a Portrait of the Female Writer as a Model of Behaviour." Ateneum Philological Forum 1/2: 25-40.
—. 2015. "Constructing a Portrait of the Early-Modern Woman Writer for Eighteenth-Century Female Readers: George Ballard's Memoirs of Several Ladies of Great Britain (1752)." Sederi — Yearbook of the Spanish and Portuguese Society for English Renaissance Studies 25: 105-27. http://www.redalyc.org/articulo. oa?id=333543052006.
Luria Walker, Gina, ed. 2005. The Idea of Being Free. A Mary Hays Reader. Peterborough: Broadview.
—. 2006. Mary Hays (1759-1843). The Growth of a Woman's Mind. Aldershot: Ashgate.
—. 2013. "General Introduction". In Female Biography; or, Memoirs of Illustrious and Celebrated Women, of All Ages and Countries. Alphabetically Arranged. Vol. I, edited by Gina Luria Walker, xi-xxx. London: Pickering & Chatto.
Macaulay, Catharine. 1994 [1790]. Letters on Education. Facsimile edition. Oxford: Woodstock Books.
Mazzeo, Tilar J. 2007. Plagiarism and Literary Property in the Romantic Period. Philadelphia, PE: University of Pennsylvania Press.
LITERATURE 93
Pascoe, Judith. 2004. "'Unsex'd females': Barbauld, Robinson, and Smith." In The Cambridge Companion to English Literature, 1740-1830, edited by Thomas Keymer and Jon Mee, 211-26. Cambridge: Cambridge University Press.
Pearson, Jacqueline. 1999. Women's Reading in Britain 1750-1835. Cambridge: Cambridge University Press.
Rivers, Isabel. 2003. "Biographical Dictionaries and their Uses from Bayle to Chalmers." In Books and their Readers in Eighteenth-Century England: New Essays, edited by Isabel Rivers, 135-69. London: Continuum.
Schwoerer, Lois G. 1988. Lady Rachel Russell "One of the Best ofWomen. " Baltimore: Johns Hopkins University Press.
Spencer, Jane. 2000. Aphra Behn's Afterlife. Oxford: Oxford University Press.
Spongberg, Mary. 2002. Writing Women's History since the Renaissance. Basingstoke: Palgrave.
—. 2010. "Mary Hays and Mary Wollstonecraft and the Evolution of Dissenting Feminism." Enlightenment and Dissent 26: 230-58.
—. 2014. "The Sources of Female Biography". In Female Biography; or, Memoirs of Illustrious and Celebrated Women, of All Ages and Countries. Alphabetically Arranged. Vol. VI, edited by Gina Luria Walker, 535-44. London: Pickering & Chatto.
Staves, Susan. 2007. "'Books without Which I Cannot Write:' How Did Eighteenth-Century Women Writers Get the Books They Read?" In Women and Material Culture, 1660-1830, edited by Jannie Batchelor and Cora Kaplan, 192-211. Basingstoke: Palgrave.
Sweet, Rosemary. 2004. Antiquaries: The Discovery of the Past in Eighteenth-Century Britain. London & New York: Palgrave.
Todd, Janet. 1998. The Critical Fortunes of Aphra Behn. Columbia SC: Camden House.
Ty, Eleanor. 1998. Introduction to The Victim of Prejudice, by Mary Hays, 2nd ed., edited by Eleanor Ty, ix-xxxix. Peterborough, Ontario: Broadview.
VV. AA. 1780. Biographia Britannica: Or, The Lives of the Most Eminent Persons Who Have Flourished in Great Britain and Ireland, from the Earliest Ages, down to the Present Times, vol. II. London: W. and A. Strahan.
Wollstonecraft, Mary. 1999. A Vindication of the Rights of Woman. A Vindication of the Rights of Men, edited by Janet Todd. Oxford: Oxford University Press.
Wood, Jeanne. 1998. "'Alphabetically Arranged': Mary Hays's Female Biography and the Biographical Dictionary." Genre 31: 117-42.
94 Begoña Lasa Alvarez Mary Hays, an Eighteenth-century Woman Lexicographer at the Service of "the Female World"
Nataša Puhner Mateja Dagarin Fojkar
University of Ljubljana, Slovenia
2018, Vol. 15 (2), 97-113(120) E revije.ff.uni-lj.si/elope
doi: 10.4312/elope.15.2.97-113 UDC: [808.1:37.091.3]:[811.111'243:373.3]
Developing EFL Writing with Year 5 Pupils by Writing Summaries
ABSTRACT
Developing EFL writing skills in primary school pupils is one of the most demanding and time consuming challenges that foreign language teachers face. The main aim of this study was thus to analyse 64 written summaries of Year 5 pupils in terms of content, structure, length and most often used words. In order to achieve the aim of pupils being able to write a summary of a story individually and in a given time frame, the task was approached gradually and systematically throughout the school year. The results show that most of the pupils managed to finish a structurally organized summary and most of them wrote a summary which included the most important aspects regarding the content of the story. The results of the survey indicate that a systematic approach to writing summaries helps pupils in developing their writing skills.
Keywords: English as a foreign language; primary school pupils; summary; writing skills
Razvijanje pisnih spretnosti v tujem jeziku pri petošolcih s pisanjem obnov
POVZETEK
Razvijanje pisnih spretnosti v angleščini kot tujem jeziku pri osnovnošolskih učencih je eden izmed najbolj zahtevnih in časovno obsežnih izzivov, s katerimi se soočajo učitelji tujih jezikov. Glavni namen te študije je bil analizirati 64 pisnih obnov petošolcev glede na vsebino, strukturo, dolžino besedila in najpogosteje uporabljene besede. Za dosego cilja, pri katerem so učenci sposobni individualno napisati obnovo v določenem času, je bil potreben postopen in sistematičen pristop k pisanju skozi vse šolsko leto. Rezultati kažejo, da je večini učencev uspelo v celoti napisati strukturno organizirano obnovo in da je večina napisala obnovo, v kateri so bili upoštevani najpomembnejši vsebinski vidiki zgodbe. Rezultati raziskave nakazujejo, da sistematičen pristop k pisanju obnov pozitivno prispeva k razvijanju pisnih spretnosti.
Ključne besede: obnova; osnovnošolski učenci; pisna spretnost; tuji jezik angleščina

ENGLISH LANGUAGE AND LITERATURE TEACHING 97
Developing EFL Writing with Year 5 Pupils by Writing Summaries
1	Introduction
English language learning is starting at an earlier age across the globe. In Slovenia compulsory English instruction now starts at the age of seven and, for at least the first couple of years focuses primarily on developing listening and speaking, while reading and writing skills are gradually introduced in later years. However, The National Examinations Centre's (RIC 2014; RIC 2015; RIC 2016; RIC 2017) conclusions over the past four years point out that the writing skills of Year 6 pupils are not achieving an adequate level according to the national syllabus for English (2016). This could be the case because national syllabuses for teaching foreign languages are often designed with no reference to any results based on literacy research and therefore offer little support to teachers of foreign languages (Dagarin Fojkar, Sesek, and Skela 2011, 19). There are no step-by-step guidelines on how to approach writing systematically and teachers are left to their own devices to construct appropriate activities and support pupils in this complex process. Consequently, teaching literacy skills is left to chance, and is a by-product of learning foreign languages (Dagarin Fojkar, Sesek, and Skela 2011, 19). In Slovenia, the development of foreign language literacy skills in primary school has not been carried out in a sufficiently systematic way (Dagarin Fojkar 2014), leading to considerable differences among pupils' writing skills.
2	Theoretical Background
"Research on the writing process, the most complex cognitive domain of language, has led to what is now the dominant paradigm in writing instruction, the process approach" (Cushing Weigle 2014, 226). Brown (2001, 320) states that "a few decades ago writing teachers were mostly concerned with the final product of writing and what that product should look like". Learners were required to write compositions in a set style following accurate grammar and conventional text organization (Brown 2001). In recent years, however, teachers have encouraged learners to create their own texts, focusing on content and message, and thus putting learners' intrinsic motives as the starting point of learning (Brown 2001). According to Shih (1986), process approaches are of an intrinsic nature because they: focus on the process of writing and do not put the final product at the centre of learning; help learners to understand the writing process; help them with strategies for prewriting, drafting, and rewriting; give learners ample time to write; emphasise the process of revision; offer learners feedback throughout the entire writing process and provide learners individual help during the writing process.
Hirvela (2016) provides a convincing rationale for including reading as an integral component of the writing classroom. One of his main areas of source-based writing is reading for writing, which is divided in the input (reading) and the output (writing) orientation. "Reading quite clearly provides invaluable input for helping individuals visualize what writing looks like in the target language" (Hirvela 2016, 48). Brown writes (2001, 330):
Clearly students learn to write in part by carefully observing what is already written. That is, they learn by observing, or reading, the written word. By reading and studying a variety of relevant types of text, students can gain important insights both about how they should write and about subject matter that may become the topic of their writing.
98 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E L O P
Stotsky (1995, 773) summarizes the benefits of reading texts before writing them by observing that: "Reading experience would seem to be the chief source of a developing writer's syntactic, generic, and lexical knowledge". Moreover, in line with Hirvela (2016, 49), "reading also provides writers with essential material to write with and about; the output orientation concerns the writing that follows the reading".
Summary writing is one of the most obvious activities after reading, despite not being a simple task. Cho (2012) claims that most pupils have difficulties writing summaries in their first language, and even more in a foreign language and although foreign language teachers see summary writing as an important skill, they lack the knowledge of how to teach it.
Many authors (Friend 2001; Westby et al. 2010; Langan 2015) have offered their definitions of a summary. Behrens and Rosen (2012, 3) define it as follows: "By summary we mean a brief restatement, in your own words, of the content of the passage (a group of paragraphs, a chapter, an article, a book). This restatement should focus on the central idea of the passage." Kirkland and Saunders (1991, 105) state that summarizing is "a highly complex, recursive reading-writing activity", and as such cognitively very demanding for learners. According to Cho (2012), the most important skill in summary writing is establishing the key ideas in the text, although Rinehart and Thomas (1993) emphasize the importance of reflection and decision making in the process of summary writing. Regarding the process rules of summary writing Hill (1991) lists the following components: text difficulty and organisation, comprehensibility and availability of text, intended audience and the purpose of writing, genre and text length. Brown and Day (1983) add the following: deletion of unimportant and redundant information, the use of hypernyms and careful selection of a topic sentence.
Johnson (1983) proposes six steps in writing story summaries. The first is understanding the text ideas, followed by connecting these ideas, identifying the text structure and remembering the content of the story. The last two steps are related to the actual writing of the story, i.e. selecting which information to include in the writing and formulating a coherent text (Johnson 1983). According to Brown and Day (1983) children are capable of excluding unimportant information at quite an early age, provided that the material they are summarizing is age-appropriate. Planning, which is a metacognitive process, plays an essential role in writing summaries. In fact, Hidi and Anderson (1986, 481) argue that "planning improved younger children's performances, and that adequate planning, which gradually emerges in development, is of central importance in summarizing."
Research on developing children's writing skills in English as a foreign language is much scarcer than research on developing children's reading or speaking skills (Murphy 2017). Lindgren and Stevenson (2013) examined the use of interactional resources in letters to a pen friend among Swedish 11-year-old pupils in their first language and in English as a foreign language. The main aims of their research were to investigate whether the language of the letters and the gender of the writers had an impact on their expression of interactional meanings, and to analyse the resources that the pupils employed to express interactional meanings in English. The results showed that there were not many differences in the frequency of interactional meanings in the two languages. However, there were a few more differences found in relation to gender, with girls using more interactional meanings than boys and writing longer letters than boys. The texts were significantly longer in the first language than in the foreign language. The pupils used many non-linguistic resources to express the meanings in English, due to the complexity of writing in a foreign language.
ENGLISH LANGUAGE AND LITERATURE TEACHING 99
Another study, conducted by Bae and Lee (2012) explored the enhancement of English writing skills of 42 pupils who participated in an extensive EFL programme over the course of 18 months, receiving 315 additional lessons. The pupils were aged from 9 to 13 and their English writing skills were tested three times during the programme. The measured subskills were grammar, content, coherence, text length and spelling. During the course of the study, the pupils consistently progressed in grammar, coherence and text length, with text length improving the fastest. The progression of spelling and content was not linear, but both still showed improvement at the final testing. The study has significance in developing writing skills in a foreign language, proving that systematic and longitudinal focus on developing writing skills brings positive results.
In their research, Perez-Paredes and Sanchez-Tornel (2014) examined the use of adverbs by learners in grades 5, 6, 9 and 10. Their aim was to establish whether the use of adverbs increases with age. The data was taken from the Polish, Spanish and Chinese Corpus of Crosslinguistic Interlanguage (ICCI) with the focus on 'food' and 'money' topics. Their study showed a significant increase in the use of adverbs with the age, and concluded that pupils up to grade 9 are less likely to use adverbs.
In her master's thesis, Zupan (2017) researched the main characteristics of 110 English texts written by Year 6 pupils at the Slovene national exam in the school year 2013/14. They were analysed according to text length, lexical density, the most common mistakes, the most common two- and three-word phrases and readability. The 50 most often used words and word groups were compared to the corpus New General Service List (New-GSL) and CEFR (Common European Framework of Reference) language levels. The results showed that an average text was 72.06 words long, the lexical density was 10.20% (Ure's method1), and the most common mistakes were grammatical. The 10 most often used words were 'in', 'be', 'I', 'you', 'a', 'can', 'and', 'the, 'do' and 'town'. Overall, 78.8% of the first 50 most often used words were also listed in the first 500 words in the corpus New-GSL; 90.4 to 96.2% words were level A1 according to CEFR, the rest were level A2. The research also showed that among the top 50 most often used words the most common word groups are nouns (28.8%), verbs (19.2%), pronouns (13.5%), and prepositions (13.5%). The most seldom used word groups were adverbs (9.6%), adjectives (5.8%), conjunctions (5.8%) and particles (3.8%). The most often used two-word phrases were 'in the', 'live in' and 'it is', and three-word phrases 'I live in', 'a lot of' and 'in your town'. The Gunning Fog readability index was 3.8 on average, which is a slightly higher result than expected, as the pupils had been formally learning English for approximately three years. This study is the most relevant for the current research, since it was conducted in the same setting and used similar tools for text analysis.
3 The Study
The aim of this study was to analyse the written summaries of Year 5 pupils in terms of content, general concepts of coherence, length and most often used words.
3.1 Research Questions
1.	How many pupils can write a summary in the limited time available following and considering the content structure (the introduction, main body and conclusion)?
2.	What is the average length of pupils' summaries?
1 Ure's method to measure lexical density: lexical density = (the number of lexical items x 100): the total words (Ure 1971). 100 Natasa Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E L O P
3.	Which words and word classes do pupils use most often?
4.	To what extent do most frequently used words in summaries correspond to the New General Service List?
5.	At which CEFR2 level are the first 50 most often used words in the summaries?
6.	What is the readability level of the summaries according to the Gunning Fog Index?
3.2 Method and Participants
3.2.1	Method
The research questions are quantitatively oriented, as they require data collection, analysis and interpretation to prove the main aim of this study. Several tools for analysing texts were used: Text Inspector, LancsLex and Textalyser. Text Inspector is the professional web tool for analysing texts. LancsLex is a tool that "analyses lexical coverage of texts and compares it to the New General Service List that identifies 2,490 most frequent words in the English language" (Brezina 2014). The tool examines lexical diversity and text complexity. Textalyser is an online text analysis tool which offers detailed statistics of the text, such as text length, lexical density and frequencies of words.
While there are numerous lists of frequent lexical items in English, the most used is West's General Service List (GSL). To improve the existing GSL List, a New General Service List was developed. It is based on four language corpora (LOB, BNC, BE06 and EnTenTen12) and consists of 13 billion running words (Brezina and Gablasova 2015).
The Gunning Fog Index is a readability test for texts written in English. It estimates the years of formal education a person needs to understand the text on the first reading (Tuckyta, Risagarniwa, and Sopian 2017). The formula for Gunning Fog is (Gunning 1952):
Reading Level (Grade) =	(Average number of words in sentences + Percentage of
words of three or more syllables) x 0.4
where complex words are defined as those containing three or more syllables. For example, technical documents usually reach a Fog Index between 10 and 15, while professional prose seldom exceeds 18 (Tuckyta, Risagarniwa, and Sopian 2017).
3.2.2	Participants
A total of 64 Year 5 pupils participated in this study. The pupils were 10 or 11 years old. By the time of the study, the pupils had completed 280 lessons of English language learning from Year 1 to Year 4. Prior to commencing the study, consent was sought from their parents or legal guardians.
3.3 Study Background
The process of developing summary writing skills already started at the beginning stages of learning English as a foreign language, by exposing pupils to stories, which offered a rich source
2 The Common European Framework (CEFR) of six broad levels gives an adequate coverage of the learning space relevant to European language learners: Breakthrough (A1), Waystage (A2), Threshold (B1), Vantage (B2), Effective Operational Proficiency (C1), Mastery (C2). "A" covering the basic user, "B" the independent user and "C" the proficient user.
ENGLISH LANGUAGE AND LITERATURE TEACHING 101
of language. In the later stages when pupils were able to detect the key words of the story, the teacher helped them to organize the words into meaningful sentences and then into an organized text. This process took four years, and throughout all this time it was closely supported by the teacher. In the first years, all summaries were written by the teacher with the help of the students. In Year 4 pupils gradually started writing their own summaries, first in pairs and then individually. Before writing the summaries, they practiced the phrases and structures appearing in the story orally through various activities, such as role-play and only later on started with the writing tasks. These needed to be more guided at first, e.g. putting the story in the correct order, gap fill and matching activities. Answering the questions in full sentences in the written form and describing the pictures were final activities before writing the whole summary.
3.4 Study Context
The aim of developing writing skills in the present study was that Year 5 pupils would be able to write a summary of a story individually and in a given time frame. The task was approached gradually and systematically throughout the school year. The pupils were faced with a challenge of writing summaries of three different stories. To enhance writing skills by writing summaries was a decision based on the still relatively limited vocabulary and very basic grammar knowledge of the pupils at this level/age, and by reading the story first the pupils were provided with the context, necessary vocabulary and grammar structures as support. The teacher's approach in this study was story-based; the pupils did not use course books.
The first summary writing was based on the book 101 Dalmatians re-told for young learners by Marie Crook (Penguin Kids, Level 3, 600 headwords, CYLET: Starters). As motivation and to enhance comprehension, pupils first watched the animated film 101 Dalmatians produced by Walt Disney in 1961 and based on the novel The Hundred and One Dalmatians by Dodie Smith. Afterwards, they read the book and did a number of activities on the worksheets prepared by their teacher to expand vocabulary and improve story comprehension, for example, marking the statements as true or false, putting the story in the correct order, matching parts of the sentences, gap fill exercises and answering the questions related to the story. Two significant processes for writing a summary had already been introduced in Year Four, namely the use of keywords and the writing of a mind map. The process of writing a draft was introduced with this story.
Prior to composing the draft the pupils listed the keywords of the story, which the teacher wrote on the board to be available to everybody. The first summary writing task was done in pairs in order to create a safe and nurturing learning environment for the pupils with ample time to complete the task. One lesson (45 minutes) was dedicated solely to the composition of the draft, in which the division of the summary into introduction, main part (body) and conclusion should have been transparent. In addition, the pupils could use all the learning materials which were available to look up words or even copy some sentences while writing the draft, which relieved them of achievement pressure. The aim was to focus on the process of writing, where there is no need for fear of writing, where mistakes are an opportunity to enhance writing abilities and not something to be judged by. The decision to use the available support had to be made before they started working on the draft, and they had to make a simple note of no help or with help on their writing papers.
The next step was writing the summary of the story using their draft. The writing time was limited to 2 lessons. Afterwards, the teacher rewrote the pupils' summaries (in a typed manner) without their errors. The pupils compared their writing with the corrected teacher's version and
102 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E L O P
tried to find their mistakes. As the next step, the pupils completed a self-assessment sheet, which was later talked through individually with the teacher who also supplied additional feedback on their writing. The pupils copied the corrected version of their summaries in their notebooks. The self-assessment sheet enabled the pupils to reflect on their writing; regarding the structure (introduction, main body, and conclusion), the content (the key events of the story), vocabulary, and intelligibility at a sentence level. In order to make the content criteria comprehensible for the pupils at this age to check, the key events of the story were written as content guidelines on the self-assessment sheet, and the pupils ticked them off as they were assessing their summaries.
The second story was Nail Soup. As an introduction and motivational tool, the pupils played a blind-folded guessing game, trying to recognize some of the key items from the story by touching them (an onion, a potato, a carrot, a pot). Listening to an authentic recording of the story (Bedtime Stories 2004) followed. The teacher checked the pupils' comprehension of the story by encouraging them to tell a brief summary. Basic story vocabulary was reinforced with a worksheet. The second listening to the story was accompanied by simultaneous reading of the printed version of the story. As a result, the pupils all confirmed their comprehension of the story to be much better on their second attempt. A number of activities and worksheets prepared by their teacher followed. The first activity was rapid word recognition, which was redone during the last lesson of the topic with a much higher rate of success than the first time. Among the activities were also classifying the words into semantic groups, putting the story into the correct order (two different versions) and other exercises similar to the ones in the previous story, yet linguistically more challenging than in the course of the first story. The next step towards pupils becoming "autonomous writers" was the individual writing of the second summary in a shorter time, i.e. in one lesson. The time frame intended for the draft stayed the same (one lesson). To further support the safe and nurturing learning environment, pupils could still use all learning materials when writing the draft, but this time only to look up words. Surprisingly, when the choice of using learning materials as a writing aid was offered to the pupils, many of them saw it as a challenge to try to write without the use of such materials, and were very keen to try to rely on their memories. The same steps as in the case of the first summary followed: teacher's correction of the summaries, pupils' comparison of the original and the corrected summary, pupils' self-assessment, teacher's individual feedback on their writing, and pupils' rewriting the corrected version.
The third story was San Francisco Story by John Escott (Penguin Readers, Easystarts, 200 headwords). As an introduction, the pupils listened to and sang two songs from different decades about San Francisco, the classic San Francisco by Scott McKenzie (1967) and a more recent one: Save me San Francisco by Train (2009). In continuation, the pupils watched two short documentaries about the earthquake in San Francisco: 1906San Francisco Earthquake (https://www.youtube.com/watch?v=-gebK-F4D1k) and the 1906— San Francisco Earthquake—Eyewitness Account by Jack London (https:// www.youtube.com/watch?v=1uhDTdQ6AWA) to get familiar with the theme. First, the pupils silently read the story on their own. In pairs, they read the story again to each other, allowing time for improving pronunciation and checking vocabulary comprehension. This was again followed by a number of writing tasks and worksheets prepared by their teacher. The tasks were similar to the ones in the previous two stories, but this time the teacher offered less support, so the pupils were expected to work autonomously. Before writing the summary, the revision of the vocabulary and phrases from the story was carried out in three stages using the mini white boards: translating words and phrases from Slovene to English, translating sentences from Slovene to English and answering the questions in full sentences. As previously, summary writing was an individual task, but the pupils could not use any materials to help them with their writing, and the writing time
ENGLISH LANGUAGE AND LITERATURE TEACHING 103
for the draft and the summary was limited to one lesson only. The same steps as in the case of the first and second summaries followed: teacher's correction of the summaries, pupils' comparison of the original and the corrected summaries, pupils' self-assessment, teacher's individual feedback on their writing, and pupils' rewriting the corrected versions. The third summary was seen as the final step towards becoming an independent summary writer in a foreign language in Year 5. As one of the main aims was to prove to the pupils that they could become confident and effective users of written language, the grammatical and spelling aspects of their writings were intentionally not included in the analysis. The grammar and spelling mistakes were corrected but no specific attention was drawn to them. Every individual summary was discussed with the author with regard to text organization, content and mistakes.
As explained above and briefly summarized in Table 1 below, the level of difficulty increased gradually and systematically regarding different types of grouping, use of learning materials while writing, and the time frame.
Table 1. A systematic increase of the level of difficulty in summary writing
STORY	PAIR or INDIVIDUAL WRITING	USE OF LEARNING MATERIALS while writing the draft	TIME FRAME	
			DRAFT	SUMMARY
101 Dalmatians	pair writing	allowed	1 lesson	2 lessons
Nail Soup	individual writing	allowed, but only to look up words	1 lesson	1 lesson
San Francisco Story	individual writing	not allowed	1 lesson	
The present study is based on the last summary, i.e. San Francisco Story, which was written individually in one lesson, without any additional help.
4 Results
The results are presented in six sections according to the main research areas.
4.1 Completion of Summary, Content and Structure
As seen in Table 2, most of the pupils (73.4%) managed to finish a structurally organized summary. Only 6.3% of those who had written a finished summary did not consider the general concepts of coherence, and did not divide their writing into an introduction, main body and conclusion paragraph. Regarding the content only, 79.7% of the pupils wrote a summary which included the most important aspects of the story. 15.6% of pupils were still not able to finish the summary in the time given. All but one pupil (who did not understand the whole story) noted not having enough time as the reason for not finishing the summary. Only 3.1% of pupils did not try writing in English but chose Slovene instead, and just one pupil (1.6% of the total) wrote an incomprehensible summary in English, one that had too many language and content-related
104 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E L O P
inadequacies to be understood. The incomprehensibly written summary and the two summaries written in Slovene were excluded from further analysis.
Table 2. Summary writing, N=64.
Incomprehensibly written summary	Summary written in Slovene	Unfinished summary	Finished, structurally not organized summary	Finished, structurally organized summary
1 (1.6%)	2 (3.1%)	10 (15.6%)	4 (6.3%)	47 (73.4%)
We were pleasantly surprised that 50.8% of all pupils who wrote a comprehensible text in English included all 13 content guidelines in their summaries. The guidelines were in the form of 13 consecutive statements about the key events and facts in the story, which should have been included in the summary in order to meet the criteria for a coherent text. The content guidelines of San Francisco Story were: 1. Mr. Tyson owns a shop in San Francisco. 2. Ben needs a job. 3. Mr. Tyson offers Ben a job and a place to stay. 4. Mr. Tyson has a daughter called Lucy. 5. Jed is jealous because Lucy likes Ben. 6. Jed steals some money and leaves it in Ben's room. 7. Mr. Tyson fires Ben thinking he is a thief. 8. Ben sleeps in a park. 9. There is an earthquake. 10. Jed steals some money again. 11. Jed does not help Lucy during the earthquake, but Ben does. 12. Lucy and Mr. Tyson find out Jed is the thief. 13. Mr. Tyson, Lucy and Ben stay together in San Francisco. In the vast majority of summaries (80.3%) ten or more guidelines were considered. Less than a half of the guidelines were included in the text in only 8.2% of the summaries.
Table 3. Content adequacy of the summaries, N=61.
Number of guidelines	Number of summaries
included in the summary	
13 (all)	31 (50.8%)
12	9 (14.8%)
11	4 (5.6%)
10	5 (8.2%)
9	2 (3.3%)
8	1 (1.6%)
7	3 (4.9%)
6	2 (3.3%)
5	3 (4.9%)
4	1 (1.6%)
Table 4. The three most often included guidelines, N=61.
Position of the guideline in the story sequence	The most often included guidelines	Number of inclusions
1	Mr. Tyson owns a shop in San Francisco.	61 (100%)
2	Ben needs a job.	60 (98.4%)
6	Jed steals some money and leaves it in Ben's room.	58 (95.1%)
ENGLISH LANGUAGE AND LITERATURE TEACHING 105
Table 5. The three least often included guidelines, N=61.
Position of the guideline in the story sequence	The least often included guidelines	Number of inclusions
10	Jed steals some money again.	43 (70.5%)
12	Lucy and Mr. Tyson find out Jed is the thief.	45 (73.8)
13	Mr. Tyson, Lucy and Ben stay together in San Francisco.	47 (77%)
4.1.1 Omission of Direct Speech
Since the book San Francisco Story is categorized as part of the 'Easystarts' series, the use of direct speech in the book is very common. One of the key summary writing skills is learning to omit direct speech, and most of the summaries (83.6%) were indeed written without any of this.
4.2 The Length of the Summaries
The average summary was composed of 15.1 sentences and the average sentence consisted of 7.7 words, which resulted in 119.8 words being written in the average summary and 57.3 of these were unique tokens (words) as individual occurrences of a linguistic unit in speech or writing.
Table 6. Sentence count, sentence length, word count and unique tokens, N=61
	Average summary
Sentence count	15.1
Sentence length in words	7.7
Word count	119.8
Unique tokens	57.3
4.3 Most Often Used Word Classes
As expected, nouns (3,179 in total) were the most often used word class in the summaries, followed by verbs (1,373 in total) and determiners/articles (627 in total). Modals (only 27 in total) and possessive pronouns (29 in total) were the least often used.
Table 7. Most often used word classes in summaries, N=61.
Word class	Number in total
Noun	3179
Verb	1373
Determiner, article	627
Preposition/subord. conj.	500
Personal pronoun	323
Adjective	311
Coordination conjunction	224
Adverb	209
106 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
Cardinal number	66
Determiner	57
Possessive pronoun	29
Modal	27
E
L
□
P
4.4	Most Frequently Used Words in Summaries in Correspondence with the New General Service List
As Table 8 shows, the five most frequently used words in the summaries coincide with the first seven words ranked by the New-GSL (rank 3, 6, 2, 1 and 7). However, regarding the content of the book, it was to be expected that some frequently used words in the summaries would not be found in the first 500 words of the New-GSL but much later on the list of the first 2500 core vocabulary words, like "store" (rank 1938) and "safe" (rank 1109), both used as nouns. The words "earthquake" and "jealous" which take 23rd and 35th place on the list of most frequently used words in the analysed summaries, are not listed in the New-GSL.
4.5	CEFR Level of the First 50 Most Often Used Words in the Summaries
Of the 50 most often used words in the summaries, 45 (90%) are level A1, one is level A2 (offer — verb), three are B1 (store — noun, thief, jealous) and one B2 (earthquake). This result was partly expected, since the book is categorized as level A1 according to the CEFR criteria.
Table 8. Most frequently used words in summaries in correspondence with the New General Service List and CEFR level of the 50 most often used words in the summaries.
Occurrence rank	Word	Occurrence frequency	Rank by	CEFR
in summaries		in summaries	New-GSL	
1	and	323	3	A1
2	a	315	6	A1
3	be	311	2	A1
4	the	282	1	A1
5	in	214	7	A1
6	he	207	11	A1
7	money	171	264	A1
8	store (noun)	139	1938	B1
9	job	118	307	A1
10	to	109	7	A1
11	have	93	8	A1
12	help	79	185	A1
13	take	79	46	A1
14	safe (noun)	71	1109	A1
15	need	68	117	A1
16	not	67	21	A1
17	thief	67	166	B1
ENGLISH LANGUAGE AND LITERATURE TEACHING 107
18	bag	64	1337	A1
19	small	60	166	A1
20	day	55	95	A1
21	one	55	36	A1
22	daughter	51	934	A1
23	earthquake	51	off list	B2
24	work	49	122	A1
25	from	48	24	A1
26	room	47	289	A1
27	say	45	32	A1
28	do	44	22	A1
29	it	42	12	A1
30	sleep	42	1507	A1
31	g°	41	47	A1
32	like	41	242	A1
33	run	41	200	A1
34	park	41	1483	A1
35	jealous	41	off list	B1
36	because	37	96	A1
37	into	37	53	A1
38	there	35	129	A1
39	put	34	150	A1
40	but	33	23	A1
41	that	32	9	A1
42	some	31	48	A1
43	stay	31	442	A1
44	offer	29	285	A2
45	give	27	74	A1
46	now	27	72	A1
47	together	27	238	A1
48	think	25	91	A1
49	know	24	61	A1
50	man	23	105	A1
4.6 The Readability Level of the Summaries according to the Gunning Fog Index
The average Gunning Fog Index is 4.11.
5 Discussion
The present study demonstrates that young learners (Year 5, age 10 or 11) can write a summary of a story individually and in a given time frame. The emphasis was on a systematic and gradual approach of developing writing skills, which was achieved by multiple summary writings during one school year.
108 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E L O P
Encouragingly, most of the pupils (73.4%) managed to finish a structurally organized summary and only 15.6% of the pupils were not able to finish it in the given time frame. Only 6.3% of all the pupils wrote a complete summary without considering the general concepts of coherence (the introduction, main body and conclusion). Just 3.1% of the pupils did not feel confident enough to write in English, so they chose Slovene instead, and only 1.6% of the pupils wrote an incomprehensible summary in English. One of the more significant findings to emerge from these results is that pupils can master key (summary) writing skills when the teacher provides them with direct instruction and practice over a longer period of time.
Regarding the content, over a half of the pupils (50.8%) who wrote a comprehensible text in English included all 13 content guidelines in their summaries. Ten or more guidelines were considered in more than 80% of the summaries. The three most often included guidelines, with a percentage of inclusion ranging from 100 to 95.1%, were from the first half of the story, the first guideline being the one with a 100% percentage inclusion. The three least often included guidelines, with a percentage of inclusion ranging from 70.5 to 77%, were from the last third of the story. Discussing the key events of the story and the order in which they happened before writing a summary helped pupils to include most of them in their writing. However, as the results indicate, the first events in the story were more memorable for the pupils than the events that happened towards the end. With regard to the omission of direct speech, the vast majority of the summaries (83.6%) were indeed written without any use of it, again due to the fact that the form of reporting events in the story was addressed during the lessons.
Further analysis showed that the average summary was composed of 15.1 sentences and the average sentence consisted of 7.72 words, which resulted in 119.8 words being written in the average summary and 57.3 of these were unique tokens. In comparison to the results of the study conducted by Zupan (2017), where the average length of the written composition of Year 6 pupils at the National Exam in English was 72.1 words, Year 5 pupils in the present study wrote texts which were 47.73 words longer. Lindgren and Stevenson (2013) examined Swedish pupils' (aged 11 or 12) writing and determined that they wrote texts with 281.5 characters on average, whereas Slovene Year 5 pupils in the present study included 459.3 characters in an average text, and thus used 177.8 (38.7%) more characters in their summaries compared to the Swedish pupils. Bae and Lee (2012) measured the text length three times during their study. In comparison to Slovene Year 5 pupils with 119.8 words per average summary, South Korean pupils (aged 9 to 13) wrote longer texts at each measurement: 125.9 words at the first measurement, 191.9 words at the second measurement, and 267 words at the third measurement; which shows that South Korean pupils already outdid Slovene pupils in their initial attempts. The reason for this improvement from the first to the third measurement could be found in the intensive learning process used in the South Korean study: a total of 315 hours of instruction just over the 18-month period. In comparison with the Slovene pupils, most of the South Korean pupils in the study were 2-3 years older and had a much larger exposure to the English language than their Slovene counterparts. In another study, in which the participants were 5, 6, 8, and 9-graders (aged between 11 and 16) from different countries, Perez-Paredes and Sanchez-Tornel (2014) established that Chinese learners wrote the largest number of words per essay (99.9), followed by Polish (87.2) and Spanish students (75.1). Comparing the two studies, Slovene Year 5 pupils wrote longer texts than any of the nations included in Perez-Paredes and Sanchez-Tornel's (2014) research. The Slovene summaries were 16.6% longer than the Chinese texts, 27.2% longer than the Polish texts, and even 37.3% longer than the Spanish texts. The results can be explained by the systematic approach to developing writing skills which was utilized in the present study.
ENGLISH LANGUAGE AND LITERATURE TEACHING 
109
The results of the current study showed the expected frequency regarding the use of word classes: nouns (3,179 in total) were the most frequently used word class in the summaries, followed by verbs (1,373) and determiners/articles (627). Modals (only 27) and possessive pronouns (29) were the least often used word classes from the list. Comparing the word classes by frequency between the writing compositions of Year 6 pupils at the National Exam in English (Zupan 2017) and the summaries written by Year 5 pupils, it can be established that the same word classes take the first two places, i.e. nouns and verbs. The frequencies of the other word classes in the writing compositions of Year 6 pupils at the National Exam in English (Zupan 2017) were in the following order: pronouns, prepositions, adverbs, adjectives, conjunctions, and articles. In contrast in the summaries the order was: determiners/articles, prepositions/subordination, conjunctions, pronouns, adjectives, coordination conjunctions, and adverbs. It is interesting to note that articles are very high on the list for Year 5 pupils but in the last place for Year 6 pupils. This could be due to the emphasis placed on the use of articles by the teachers during lessons. In our case, the teacher consistently used the articles in the English class herself and encouraged the pupils to use them as well, first in the form of simple repetition and then step-by-step explanation of the correct use of the articles in an age appropriate manner. Stories are a great source of texts which introduce the use of the articles in a meaningful context. Moreover, there is little emphasis on explicit teaching of the pronouns in English classes up to Year 6, which could explain the discrepancy in the use of pronouns between the pupils in Zupan's study (2017) and the present work.
The five most frequently used words in the summaries ("and", "a", "be", "the", "in") coincide with the first seven words ranked by the New-GSL (rank 3, 6, 2, 1 and 7). Comparing the results with Zupan's (2017) research, in which the first five most frequently used words were "in", "be", "I", "you", and "a", three of the five words correspond to the five most frequently used words in the summaries in the present study. Both studies confirm the high frequency of the above mentioned words and are in accordance with the New-GSL ranking. Some frequently used words in the summaries were not found in the first 500 words of the New-GSL but much later, on the list of the first 2500 core vocabulary words, like "store" (rank 8 in the summaries, rank 1938 in the New-GSL) and "safe" (rank 14 in the summaries, rank 1109 in the New-GSL) both used as nouns. The words "earthquake" and "jealous", which took 23rd and 35th places on the list of the most frequently used words in the summaries, are not listed in the New-GSL. These words are not very frequent in the every-day context, but they appear in the book San Francisco Story as keywords, due to the content of the story. Consequently, the pupils were exposed to them very often in the classroom, and thus able to use them in their summaries.
As it was expected, since the book San Francisco Story is categorized as level A1, from the 50 most often used words in the summaries 45 (90%) are level A1, one is level A2 ("offer" — verb), three are B1 ("store" — noun, "thief", "jealous") and one B2 ("earthquake") whereas in the written compositions of Year 6 pupils at the National Exam in English (Zupan, 2017) all but the last three of the 50 most often used words ("can", "swimming", "church" — all level A2) were level A1. The B1 and B2 words used by the pupils in the present study were often repeated in the book San Francisco Story, the content of which was extensively discussed prior to writing the summaries. This study shows that pupils expand their vocabulary in a natural manner through experiencing the story in a multisensory way and through doing a variety of tasks related to the content of the story. Therefore, their vocabulary level is higher than the level of the pupils who are not exposed to the stories in such an extensive way. If the words are presented in an interesting context and there is enough meaningful repetition, the pupils do not perceive these words as difficult.
110 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E
The Gunning Fog Index of the analysed Year 5 pupils' summaries is 4.11, whereas that of the Year 6 pupils' texts at the National Exam in English was 3.8% (Zupan, 2017), showing that a systematic approach to writing can result in a better readability index of pupils' written products.
6 Conclusion
The issue of teaching and acquiring writing skills in a foreign language is broad and diverse. As primarily a reading and writing activity, summary writing tasks are an effective strategy for teaching writing skills. However, despite this, most foreign language teachers rarely include the development of this skill into their instruction. Consequently, foreign language pupils cannot develop it sufficiently, despite having a large number of FL lessons at schools. Teachers in EFL classrooms should not assume that their pupils will gradually master summary writing skills in the process of learning. They need to offer their students systematic practice and direct instruction over a long period of time. Therefore, a variety of activities should be included to help students develop their (summary) writing skills. The process itself starts well before the actual writing, with activities that help pupils understand the text and remember its content. When developing summary writing skills the teacher's focus should lie on the process of writing by helping pupils to learn the most important steps of summarizing, with finding the key ideas and excluding the unimportant information being two of these. Further significant aspects teachers should bear in mind are to give pupils enough time for writing and offer them supportive feedback on their work. The results of the present study might therefore encourage foreign language teachers not to use course books alone in their classes, but also to make some extra time for extensive and in-depth reading of texts (stories in the case of early teaching of English) which enable the intensive development of writing skills in the form of summarization. Further research on developing EFL writing skills with primary school pupils is evidently an important issue which could include an analysis of other forms of writing (e.g. stories, compositions) and other models of developing processes in writing.
References
Bae, Jungok, and Yae-Sheik Lee. 2012. "Evaluating the Development of Children's Writing Ability in an EFL Context." Language Assessment Quarterly 9 (4): 348-74. https://doi.org/10.1080/15434303.2012.721424.
Behrens, Laurence M., and Leonard J. Rosen, 2012. Writing and Reading across the Curriculum. Twelfth Edition. New York: Longman.
Brezina, Vaclav. 2014. "Analyse Your Texts with the New GLS." Lancaster: Lancaster University Press. http://corpora.lancs.ac.uk/vocab/analyse.php.
Brezina, Vaclav, and Dana Gablasova. 2015. "Is There a Core General Vocabulary? Introducing the New General Service List." Applied Linguistics 36 (1): 1-22. https://doi.org/10.1093/applin/amt018.
Brown, H. Douglas. 2001. Teaching by Principles: An Interactive Approach to Language Pedagogy. New York: Longman.
Brown, Ann L., and Jeanne D. Day. 1983. "Macrorules for Summarizing Texts: The Development of
Expertise." Journal of Verbal Learning and Verbal Behavior 22 (1): 1-14. https://doi.org/10.1016/S0022-5371(83)80002-4.
Cho, Youngmyoung. 2012. "Teaching Summary Writing through Direct Instruction to Improve Text Comprehension for Students in ESL/EFL Classroom." Master's Thesis, University of Wisconsin.
ENGLISH LANGUAGE AND LITERATURE TEACHING 111
Cushing Weigle, Sara. 2014. "Considerations for Teaching Second Language Writing." In Teaching English as a Second or Foreign Language, Fourth Edition, edited by Marianne Celce-Murcia, Donna M. Brinton, and Marguerite Ann Snow, 222—37. Boston: Heinle Cengage.
Dagarin Fojkar, Mateja. 2014. "Literacy Development in Course Books for Teaching English in the Second Cycle of Slovenian Primary School." Linguistica 54 (1): 153-66. https://doi.org/10.4312/ linguistica.54.1.153-166.
Dagarin Fojkar, Mateja, Urška Sešek, and Janez Skela. 2011. Sounds and Letters. Priročnik k delovnem zvezku za razvijanje opismenjevanja v angleščini kot tujem jeziku. Ljubljana: Tangram.
Friend, Rosalie. 2001. "Effects of Strategy Instruction on Summary Writing of College Students."
Contemporary Educational Psychology 26: 3-24.
Gunning, Robert. 1952. The Technique of Clear Writing. New York: McGraw-Hill.
Hidi, Suzanne, and Valerie Anderson. 1986. "Producing Written Summaries: Task Demands, Cognitive
Operations, and Implications for Instruction." Review of Educational Research 56 (4): 473—93. https://doi. org/10.3102/00346543056004473.
Hill, Margaret. 1991. " Writing Summaries Promotes Thinking and Learning Across the Curriculum — But Why Are They So Difficult to Write?" Journal of Reading 34 (7): 536—39. http://www.jstor.org/ stable/40014578.
Hirvela, Alan R. 2016. Connecting Reading & Writing in Second Language Writing Instruction. Second Edition. Ann Arbor: The University of Michigan Press.
Johnson, Nancy. 1983. "What Do You Do If You Can't Tell the Whole Story? The Development of Summarization Skills." In Children's Language Vol. 4, edited by Keith N. Nelson, 162—63. Hillsdale, NJ: Erlbaum.
Kirkland, Margaret R., and Mary Anne P. Saunders. 1991. "Maximizing Student Performance in Summary Writing: Managing Cognitive Load." TESOL Quarterly 25 (1): 105—21. https://doi. org/10.2307/3587030.
Langan, John. 2015. Ten Steps to Advancing College Reading Skills. Sixth Edition. West Berlin, New Jersey: Townsend Press.
Lindgren Eva, and Marie Stevenson. 2013. "Interactional Resources in the Letters of Young Writers in
Swedish and English." Journal of Second Language Writing 22 (4): 390—405. https://doi.org/10.1016/j. jslw.2013.09.001.
Murphy, Victoria A. 2017. "Literacy Development in Children with English as an Additional Language (EAL)." In Early Language Learning. Complexity and Mixed Methods, edited by Janet Enever, and Eva Lindgren, 41—61. Bristol: Multilingual Matters.
Perez-Paredes, Pascual, and Maria Sanchez-Tornel. 2014. "Adverb Use and Language Proficiency in
Young Learners' Writing." International Journal of Corpus Linguistics 19 (2): 178—200. doi 10.1075/ ijcl.19.2.02per.
Program osnovna šola. Učni načrt: Angleščina. 2016. Ljubljana: Ministrstvo za šolstvo, znanost in šport,
Zavod RS za šolstvo. http://www.mizs.gov.si/fileadmin/mizs.gov.si/pageuploads/podrocje/os/prenovljeni_ UN/UN_anglescina.pdf.
RIC. 2014. "Opisi dosežkov učencev 6. razreda in specifikacijske tabele preizkusov znanja NPZ 2014." http:// www.ric.si/preverjanje_znanja/statisticni_podatki/.
—. 2015. "Opisi dosežkov učencev 6. razreda in specifikacijske tabele preizkusov znanja NPZ 2015." http:// www.ric.si/preverjanje_znanja/statisticni_podatki/.
—. 2016. "Opisi dosežkov učencev 6. razreda in specifikacijske tabele preizkusov znanja NPZ 2016." https:// www.ric.si/preverjanje_znanja/statisticni_podatki/.
112 Nataša Puhner, Mateja Dagarin Fojkar Developing EFL Writing with Year 5 Pupils by Writing Summaries
E L O P
—. 2017. "Opisi dosežkov učencev 6. razreda in specifikacijske tabele preizkusov znanja NPZ 2017." http:// www.ric.si/preverjanje_znanja/statisticni_podatki/.
Rinehart, Steven D., and Karen F. Thomas. 1993. "Summarization Ability and Text Recall by Novice
Studiers." Reading, Research and Instruction 32 (4): 24-32. https://doi.org/10.1080/19388079309558130.
Shih, May. 1986. "Content-Based Approaches to Teaching Academic Writing." TESOL Quarterly 20 (4): 617-48.
Stotsky, Sandra. 1995. "The Uses and Limitations of Personal or Personalized Writing in Writing Theory, Research, and Instruction." Reading Research Quarterly 30: 607-32.
Textalyser. 2004. http://textalyser.net/.
Text Inspector. 2015. Cambridge: Cambridge University Press. http://www.englishprofile.org/wordlists/text-inspector.
Tuckyta, Eva Sari Sujanta, Yuyu Yohana Risagarniwa, and Rahmat Sopian. 2017. The Readability Test of the English Children Short Stories. International Conference on Language, Education, Humanities and Innovation. 21st &22nd ^January, 2017: 85-92. https://proceedings.icsai.org/5iclehi/5iclehi-076.pdf
Ure, Jean. 1971. "Lexical Density and Register Differentiation." In Applications of Linguistics: Selected Papers of the 2nd International Congress of Applied Linguists, edited by George Perren and John Leslie Melville Trimm, 443-52. London: Cambridge University Press.
Westby, Carol, Barbara Culatta, Barbara Lawrence, and Kendra Hall-Kenyon. 2010. "Summarizing Expository Texts." Topics in Language Disorders 30 (4): 275-287. https://doi.org/10.1097/TLD.0b013e3181ff5a88.
Župan, Tina. 2017. "Značilnosti angleških pisnih besedil slovenskih šestošolcev." Master's Thesis, Faculty of Education, University of Ljubljana.
ENGLISH LANGUAGE AND LITERATURE TEACHING 113
w
LIST OF CONTRIBUTORS
Begoña Lasa Alvarez
Facultade de Ciencias da Educación, Universidade da Coruña Campus da Zapateira, s/n, 15071 A Coruña, Spain b.lasa@udc.es
Mateja Dagarin Fojkar
University of Ljubljana, Faculty of Education Kardeljeva ploščad 16, 1000 Ljubljana, Slovenia mateja.dagarin@pef-uni-lj.si
Nikola Dobric
Alpen-Adria-Universität Klagenfurt
Universitätsstraße 65-67, 9020 Klagenfurt am Wörthersee, Austria nikola.dobric@aau.at
\fesna Lazovic
University of Ljubljana, Faculty of Arts Aškerčeva cesta 2, 1000 Ljubljana, Slovenia vesna.lazovic@ff.uni-lj.si
Jakob Lenardič
University of Ljubljana, Faculty of Arts Aškerčeva cesta 2, 1000 Ljubljana, Slovenia jakob.lenardic@ff.uni-lj.si
Frančiška Lipovšek
University of Ljubljana, Faculty of Arts Aškerčeva cesta 2, 1000 Ljubljana, Slovenia franciska.lipovsek@ff.uni-lj.si
Nataša Puhner
University of Ljubljana, Faculty of Education Kardeljeva ploščad 16, 1000 Ljubljana, Slovenia tasamithans@hotmail.com
117
GUIDELINES FOR CONTRIBUTORS
ELOPE English Language Overseas Perspectives and Enquiries
ELOPE publishes original research articles, studies and essays that address matters pertaining to the English language, literature, teaching and translation.
Submission of Manuscripts
Manuscripts should be submitted for blind review in electronic form using the Faculty of Arts (University of Ljubljana) OJS platform (https://reviie.ff.uni-li.si/elope/about/submissions). Only one contribution by the same author per volume will be considered. Each paper should be accompanied by abstracts in English and Slovene (between 150 and 180 words each) and keywords. Abstracts by non-native speakers of Slovene will be translated into Slovene by ELOPE. Please be sure to have a qualified native speaker proofread your English-language article. Suggested length of manuscripts is between 5,000 and 8,000 words.
Manuscript Style and Format
The manuscript should be in the following format:
•	title in English,
•	abstracts in English and Slovene,
•	the text should be divided into introduction; body ofthe paper (possibly subdivided); and conclusion.
The text should be preferably written in Word format (OpenOffice and RTF files are also acceptable). Please observe the following:
•	12-point Times New Roman font size,
•	2.5 cm page margins on all sides,
•	1.5-point line spacing,
•	left text alignment,
•	footnotes should be brief (up to 300 words per page; 10-point Times New Roman font size).
For resolving practical style and formatting queries, please see the articles in the latest on-line issue (https://reviie.ff.uni-li.si/elope/issue/current) or contact the technical editor.
References
References should comply with The Chicago Manual of Style (17th edition, 2017) author-date system. A Quick Guide to CMS is available here: http://www.chicagomanualofstyle.org/tools citationguide/ citation-guide-2.html.
Final note
Please note that only manuscripts fully adhering to the ELOPE Guidelines for Contributors will be considered for publication.
E L O P
English Language Overseas Perspectives and inquiries
ELOPE Vol. 15, No. 2 (2018)
Journal Editors
Smiljana Komar, University of Ljubljana, Slovenia Mojca Krevel, University of Ljubljana, Slovenia
Editorial Board
Lisa Botshon, University of Maine at Augusta, United States of America; Michael Devine, Acadia University, Canada; Dušan Gabrovšek, University of Ljubljana, Slovenia; Michelle Gadpaille, University of Maribor, Slovenia; Meta Grosman, University of Ljubljana, Slovenia; Adolphe Haberer, University of Lyon, France; Allan James, University of Klagenfurt, Austria; Victor Kennedy, University of Maribor, Slovenia; Bernhard Kettemann, University of Graz, Austria; J. Lachlan Mackenzie, VU University Amsterdam, Netherlands; Tomaž Onič, University of Maribor, Slovenia; Roger D. Sell, Abo Akademi University, Finland; Andrej Stopar, University of Ljubljana, Slovenia; Rick Van Noy, Radford University, United States of America; Terri-ann White, University of Western Australia, Australia
Editorial Secretary
Gašper Ilc, University of Ljubljana, Slovenia
Technical Editor
Andrej Stopar, University of Ljubljana, Slovenia
Proofreading
Paul Steed
Editorial Policy
ELOPE: English Language Overseas Perspectives and Enquiries is a double-blind, peer-reviewed academic journal that publishes original research articles, studies and essays that address matters pertaining to the English language, literature, teaching and translation.
The journal promotes the discussion of linguistic and literary issues from theoretical and applied perspectives regardless of school of thought or methodology. Covering a wide range of issues and concerns, ELOPE aims to investigate and highlight the themes explored by contemporary scholars in the diverse fields of English studies.
E nglish L anguage Overseas Perspectives and inquiries
Published by
Ljubljana University Press, Faculty of Arts Znanstvena založba Filozofske fakultete Univerze v Ljubljani Aškerčeva 2, 1000 Ljubljana, Slovenia For the Publisher: Roman Kuhar, the Dean of the Faculty of Arts
Issued by
Slovene Association for the Study of English Slovensko društvo za angleške študije Aškerčeva 2, 1000 Ljubljana, Slovenia
Department of English, Faculty of Arts, University of Ljubljana Oddelek za anglistiko in amerikanistiko, Filozofska fakulteta, Univerza v Ljubljan Aškerčeva 2, 1000 Ljubljana, Slovenia
The journal is published with support from the Slovenian Research Agency.
The publication is free of charge.
Universal Decimal Classification (UDC)
Kristina Pegan Vičič
Journal Design
Gašper Mrak
Cover
Marjan Pogačnik: Zimsko cvetje, 1994 7.6 x 10.0 cm; colour etching, deep relief Owner: National Gallery, Ljubljana, Photo: Bojan Salaj, National Gallery, Ljubljana
Printed by
Birografika Bori
Cf
Number of Copies
110
doi: 10.4312/elope.15.2 Online ISSN: 2386-0316
University of Ljubljana
FACUHYofARTS	Print ISSN: 1581-8918
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.