189
DirKorp: A Croatian Corpus of 
Directive Speech Acts (v3.0)
Petra BAGO
Faculty of Humanities and Social Sciences, University of Zagreb
Virna KARLIĆ
Faculty of Humanities and Social Sciences, University of Zagreb
In this paper, we present recent developments on a new version (v3.0) of Dir-
Korp (Korpus direktivnih govornih činova hrvatskoga jezika), the first Croatian 
corpus of directive speech acts developed for the purposes of pragmatic re-
search. The corpus contains 800 elicited speech acts collected via an online 
questionnaire with role-playing tasks, a method of simulated communication 
that is implemented under pre-set conditions. This method is suitable for re-
searching speech acts due to the ability to collect a great number of examples 
of such acts of equal propositional content and illocutionary purpose used in 
the same controlled situations. The presented situations are classified into two 
categories with regard to the relationship between the participants of the com-
munication act: (1) situations involving interlocutors who are not in a familiar 
relationship; (2) situations involving interlocutors in a familiar relationship. As-
signments of the two categories are organized into four pairs, asking respond-
ents to share a speech act of similar propositional content. The respondents 
were 100 Croatian speakers, all undergraduate (63%) or graduate students 
(37%) of the Faculty of Humanities and Social Sciences (University of Zagreb). 
The corpus has been manually annotated on the speech act level, each speech 
act containing up to 14 features: (1) respondent ID, (2) familiarity/unfamiliarity, 
(3) utterance type, (4) directive performative verb in 1st person, (5) illocutionary 
force, (6) propositional content, (7) T/V form, (8) exhortative, (9) lexical marker 
of request, (10) lexical marker of apology, (11) lexical marker of gratitude, (12) 
Bago, P., Karlić, V.: DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0). 
Slovenščina 2.0, 11(1): 189–217. 
1.01 Izvirni znanstveni članek / Original Scientific Article
DOI: https://doi.org/10.4312/slo2.0.2023.1.189-217
https://creativecommons.org/licenses/by-sa/4.0/
190
Slovenščina 2.0, 2023 (1) | Articles
honorific title, (13) grammatical mood, and (14) modal verb in 2nd person. It con-
tains 12,676 tokens and 1,692 types. The corpus is encoded according to the 
TEI P5: Guidelines for Electronic Text Encoding and Interchange, developed and 
maintained by the Text Encoding Initiative Consortium (TEI). DirKorp is available 
for download under the CC BY-SA 4.0 license from GitHub in TEI format. We 
describe applied pragmatic annotation as well as the structure of the corpus.
Keywords: corpus pragmatics, directive speech acts, DirKorp, Croatian language
1 Introduction 
Corpus pragmatics is an interdisciplinary field of study that incorporates 
linguistic pragmatics and computer science, focusing on the develop-
ment of natural language corpora in machine-readable form and their 
application for the purposes of studying pragmatics phenomena in writ-
ten and spoken language. For a long time, linguists have regarded a 
corpus approach to language as incompatible with pragmatics (Rome-
ro-Trillo, 2008, p. 2). While the corpus approach to studying language 
implies processing authentic language material by implementing quan-
titative research methods, pragmatic research is still predominantly 
of a qualitative nature – based on the researcher’s introspection, data 
obtained by elicitation methods, or an analysis of authentic linguistic 
material of small size. The application of corpus analysis in the research 
of pragmatics phenomena represents a major turnaround in the devel-
opment of pragmatics, primarily because it allows a systematic analysis 
of language material of large size, and thus the detection of patterns 
of language use that “fly below the radar” through qualitative analyses 
(ibid.). In addition, it should be pointed out that the application of new 
technologies in linguistics, including pragmatics, did not only ensure, 
facilitate or accelerate numerous research processes but opened the 
door to a new, different way of thinking about language (Leech, 1992).
The application of corpus methods to large pragmatic corpora allows 
one to systematically carry out empirically based pragmatic research 
(Bunt, 2017, p. 327). While the implementation of corpus research can 
result in minor adjustments to existing theories on the one hand, it can 
lead to a rethinking of pragmatic concepts and theoretical frameworks on 
the other, such as the development of the theory of dialogue acts (ibid.).
191
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
According to Rühlemann and Aijmer (2015), one of the major 
methodological problems that corpus pragmatic researchers encoun-
ter is the disproportionate relationship between pragmatic functions 
and language forms by which these functions are expressed. One form 
can perform multiple pragmatic functions in discourse, while one func-
tion can be expressed by different forms, which makes the process of 
querying a corpus according to the pragmatic function criterion rather 
difficult. It is for this reason that corpus pragmatic researchers most 
often investigate conventional speech acts or functions performed by 
a limited number of language forms (Jucker et al., 2009, p. 4). The aim 
of this paper is to present the first Croatian corpus of directive speech 
acts, DirKorp, manually annotated for corpus pragmatic research.
The paper is structured as follows: Section 2 describes selected 
work related to corpus pragmatic research, Section 3 explores the defi-
nition, classification, and research methods of directive speech acts, 
while the subsequent three sections present the DirKorp corpus. Sec-
tion 4 gives a description of the developed corpus, Section 5 describes 
14 annotation features, and Section 6 presents the structure of the 
corpus encoded according to the TEI P5: Guidelines for Electronic Text 
Encoding and Interchange (TEI Consortium, 2021). Finally, Section 7 
contains the conclusions and some directions for future work.
This is a follow-up paper from the conference “Language Technol-
ogies and Digital Humanities” held in Ljubljana, Slovenia on 15th–16th 
September 2022, where we presented DirKorp v2.0. Here we present 
a new published version of the DirKorp (v3.0) with two additional an-
notation layers, as well as a new section clarifying the definition, clas-
sification, and research methods of directive speech acts (Section 3).
2 Related work
The number of large corpora with systematically implemented pragmatic 
annotation remains relatively small. Due to a disproportionate relation-
ship between pragmatic functions and the language forms by which 
these functions are expressed, automatic corpus annotation does not 
produce satisfactory results. For this reason, only a few researchers have 
engaged in creating larger corpora of this sort. Generally, for the purposes 
192
Slovenščina 2.0, 2023 (1) | Articles
of corpus pragmatic research, specialized corpora of smaller size are pro-
duced for individual research purposes. In addition, pragmatic research 
is sometimes carried out on corpora without pragmatic annotation.
An example of a corpus that does not contain pragmatic annota-
tion but was used for pragmatic research is the Birmingham Blog Cor-
pus1 (Kehoe and Gee, 2007; 2012). In fact, this is a subcorpus of a 
larger set of corpora being developed at the department Research and 
Development Unit for English Studies at the Birmingham City University. 
It consists of blog posts and reader comments, and includes some 500 
million words in English that were collected between 2000 and 2010. 
Automatic POS annotation was performed using the Stanford Core NLP 
tools2 and included lemma annotations and part-of-speech categories3 
based on the Universal Dependencies framework,4 while the docu-
ments contain metadata of the publication date. Pragmatic research on 
speech acts has been conducted on this corpus. For example, Lutzky 
and Kehoe (2017a; 2017b) used it to analyse apologies as speech acts 
that contain formulaic expressions, which facilitate their querying in a 
corpus when using the available tools.
Similarly, we (Karlić and Bago, 2021) conducted research on the 
pragmatic functions and properties of imperatives using corpora with-
out pragmatic annotation. We used hrWaC and srWaC (Ljubešić and 
Klubička, 2014), two large web corpora of the Croatian and Serbian 
languages with morphosyntactic annotation. For the purposes of the 
analysis, an additional pragmatic annotation of a representative sample 
of verbs in an imperative form was carried out manually. Other corpora 
of the Croatian spoken and written language with no pragmatic annota-
tion have also been used as a resource for corpus pragmatic research. 
For example, Hržica, Košutar, and Posavec (2021) used the Croatian 
Corpus of the Spoken Language of Adults (HrAL) (Kuvač Kraljević and 
Hržica, 2016) and the Croatian National Corpus of the written language 
(HNK) (Tadić, 1996) for the search and analysis of connectors and dis-
course markers.
1 https://www.webcorp.org.uk/wcx/lse/corpora
2 https://stanfordnlp.github.io/CoreNLP/
3 See more about the POS tagset used for the Birmingham Blog Corpus: https://www.webcorp.
org.uk/wcx/lse/guide.
4 https://universaldependencies.org/u/pos/index.html
193
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
According to Bunt (2017) the majority of corpora with pragmatic 
annotation contain labels on discourse relationships in written texts 
and on spoken dialogue acts. An example of such a larger corpus is the 
Penn Discourse Treebank or PDTB5 (Prasad et al., 2018) which con-
tains labels on discourse relations, i.e., discourse structure and its se-
mantics. Discourse annotations were added to a subcorpus consisting 
of texts published in the newspaper Wall Street Journal with a total 
of around 1 million tokens, included in a bigger corpus Penn Treebank 
(PTB). Bunt (2017) states that there are corpora of other languages 
developed for the purposes of studying the co-occurrence of discourse 
labels, such as Chinese, Czech, Dutch, German, Hindi, and Turkish – 
emphasizing that these corpora are manually annotated and of modest 
size. Additionally, for each corpus a new schema was developed based 
on various theoretical starting points.
DialogBank6 (Bunt et al., 2019) is one of a rare dialogue corpus anno-
tated with an ISO 24617-2 standard. It contains already existing dialogue 
corpora annotated with various schemas. Four corpora are of English, 
namely HCRC Map Task (Anderson et al., 1991), Switchboard (Godfrey et 
al., 1992), TRAINS (Allen et al., 1995) and DBOX (Petukhova et al., 2014), 
and four of Dutch – DIAMOND (Geertzen et al., 2004), OVIS7, Dutch Map 
Task (Caspers, 2000) and Schiphol (Prüst et al., 1984). Dialogue act anno-
tation involves segmenting a dialogue into defined grammatical units and 
augmenting each unit with one or more communicative function labels.
Another example of a corpus with a pragmatic annotation is the 
Engineering Lecture Corpus8 (Alsop and Nesi, 2013; 2014) which con-
tains 76 transcripts based on hour-long video recordings of engineering 
lectures held in English at three universities. It is manually annotated 
for three pragmatic features: humour, storytelling, and summary.9 Each 
feature can be augmented with one of the attributes containing addi-
tional information that describes the feature in more detail. Further, the 
corpus contains labels regarding significant breaks, laughter, writing or 
drawing on the board, etc.
5 https://doi.org/10.35111/qebf-gk47
6 https://dialogbank.lsv.uni-saarland.de/
7 http://www.let.rug.nl/vannoord/Ovis/
8 www.coventry.ac.uk/elc
9 https://www.coventry.ac.uk/research/research-directories/current-projects/2015/engineer-
ing-lecture-corpus-elc/annotations-and-mark-ups/
194
Slovenščina 2.0, 2023 (1) | Articles
Finally, we present the SPICE-Ireland corpus (Systems of Pragmat-
ic Annotation in the Spoken Component of ICE-Ireland) (Kallen and Kirk, 
2012), a part of a larger set of corpora ICE-Ireland (International Cor-
pus of English: Ireland Component) containing pragmatic, discourse, 
and prosodic features. The corpus contains various types of private 
and public, formal and informal dialogues and monologues of a length 
of about 2,000 words, with a size of some 625,000 words. It consists 
of spoken English. The pragmatic annotation of speech acts is based 
on Searle’s classification (Searle, 1969; 1976): representatives, direc-
tives, commissives, expressives, and declaratives.
When it comes to corpus research of speech acts, researchers 
have two options: (1) to analyse examples from existing corpora of au-
thentic linguistic material, or (2) to analyse examples of elicited lin-
guistic material. In the second case, different types of data completion 
tests are usually applied, and based on the obtained results smaller 
custom-made corpora are created for the needs of individual research 
(and therefore not publicly available). This method is most often used 
in cross-linguistic, contrastive research, but it is also used in the study 
of individual languages (e.g., Barron, 2008; Trosborg, 1995). For an 
overview of pragmatic research of speech acts (including directives) on 
elicited linguistic material, see, for example, Wojtaszek (2008).
To the best of our knowledge, there exist no publicly available corpo-
ra of spoken or written Croatian with pragmatic annotation. So far, Croa-
tian linguists have mostly dealt with speech acts from a theoretical per-
spective, referring primarily to the Austin’s and Searle’s theory (cf. Pupo-
vac, 1990; Ivanetić, 1995; Miščević, 2018; Palašić, 2020). However, in 
recent years the number of research projects based on the qualitative 
and quantitative analysis of small-sized authentic linguistic materials 
(from literary texts and advertisements to email messages and political 
discourse in Croatian and other languages) has been increasing (cf. e.g., 
Pišković, 2007; Matić, 2011; Franović and Šnajder, 2012; Šegić, 2019).
3 Directive speech acts: definition, classification, and 
research methods
During verbal communication, speakers express their thoughts in the 
form of utterances, through which they convey information, express 
195
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
their emotions and attitudes, or try to modify the addressee’s behav-
iour (Capone, 2009, p. 1015).
Speech acts are utterances with specific properties and communi-
cative functions (ibid.):
A speech act (…) is not merely the expression of a thought. It is the vo-
calization of a certain representation of the world (external or internal) 
aimed at making official the display of an intention to change a state 
of things and at changing things by the public display of that intention.
Therefore, speech acts can be briefly defined as “actions performed 
via utterances” (Yule, 2002, p. 47). According to Searle (1975), there are 
five types of speech acts: (1) representatives – statements that can be 
evaluated as true or false; (2) directives, which speakers use to influence 
the addressee’s wishes and actions; (3) commissives, through which 
speakers commit to perform some action in the future; (4) expressives, 
which speakers use to express their feelings or attitudes; and (5) de-
claratives or institutionalized declarations that formally change the state 
of affairs in extralinguistic reality (cf. Karlić and Bago, 2021).
Directive speech acts, or directives, are a type of speech act by 
which speakers express their “(...) desire/wish for the addressee to do 
something. (...) In using a directive, the speaker intends to elicit some 
future course of action on the part of the addressee, thus making the 
world match the words via the addressee” (Huang, 2009, p. 1004).
Directives differ with respect to their illocutionary force. The illo-
cutionary force of a directive depends on how binding it is for the ad-
dressee. If the speaker insists on its realization, the illocutionary force 
of the directive is strong – and vice versa. According to this criterion, 
directive speech acts are classified into orders, commands, requests, 
pleads, incentives, advice, etc. (cf. Piper et al., 2005, p. 1021; Karlić 
and Bago, 2021, p. 37).
Directive speech acts can be direct or indirect. Direct directives 
contain an explicit directiveness marker – an imperative (Close the win-
dow) or a performative verb in the first person of the present tense (I 
ask you to close the window). Directiveness can be expressed implic-
itly, through assertions without a performative verb (You should close 
the window), interrogative utterances (Can you close the window?), or 
196
Slovenščina 2.0, 2023 (1) | Articles
elliptical utterances (Um… the window…). Just like illocutionary force, 
the propositional content of directive speech acts can also be ex-
pressed explicitly and implicitly (It is cold here [implicature: Close the 
window]) (cf. Huang, 2009, p. 1005; Karlić and Bago, 2021, p. 39).
According to Brown and Levinson (1987, p. 65–66), directives 
represent a typical example of face-threatening acts. For this reason, 
when using them, speakers often apply various politeness strategies 
that mitigate their illocutionary force (e.g., implicatures and lexical or 
grammatical modifiers of illocutionary force).
The foundations of speech act theory were laid by the philoso-
phers John Austin and John Searle in works published in the 1960s 
and 1970s. Since then, numerous studies of speech acts have been 
conducted. In the beginning, they were non-empirical, based on the 
researcher’s intuition. In recent years, however, the number of empiri-
cal studies of speech acts has grown significantly. Jucker (2009) distin-
guishes three types of data collection methods for the needs of empiri-
cal research of speech acts – field, laboratory, and armchair (Flӧck and 
Geluykens, 2015, p. 10):
While armchair approaches investigate participants’ intuitions and at-
titudes about language use, field and laboratory approaches aim at 
studying actual language use. They differ, however, in the way lan-
guage data are produced. While in laboratory approaches, language 
use is elicited by researchers (by employing role-plays or administer-
ing discourse completion tasks), field data are defined by the absence 
of such elicitation techniques. Field methods are therefore observa-
tional in nature, i.e., they require an authentic communicative intent 
by participants to produce language.
Each of the mentioned methods has its advantages and disadvan-
tages. For the purposes of creating the DirKorp corpus, we applied the 
laboratory method of eliciting language production by role-playing. The 
main advantage of this method is that it gives “full variable control to 
the researcher (...), and can generate large amounts of data; however, 
participants use language without their own intrinsic communicative 
intent in fictional scenarios” (Flӧck and Geluykens, 2015, p. 11). This 
method allowed us to collect a large amount of mutually comparable 
197
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
directives with the same propositional content and produced in the 
same controlled circumstances.
In the following sections, we present a new version (v3.0) of Dir-
Korp, the first Croatian corpus of directive speech acts.
4 Corpus description
DirKorp (Korpus direktivnih govornih činova hrvatskoga jezika) (Karlić 
and Bago, 2021) is a Croatian corpus of directive speech acts developed 
for the purposes of pragmatic research. The corpus contains 800 elic-
ited speech acts collected via an online questionnaire with role-playing 
tasks applying the method of simulated communication that is imple-
mented under pre-set conditions. This method is suitable for research-
ing speech acts due to the ability to collect a great number of examples 
of speech acts of equal propositional content and illocutionary purpose 
used in the same controlled situations. The questionnaire included eight 
closed-type role-playing tasks. These types of tasks imply recording 
the speaker’s reactions (in this case in writing) to the stimulus without 
feedback. In each task, the participants are presented with one textually 
described hypothetical situation asking them to refer a directive speech 
act to their interlocutor. Their assignment was to imagine they were in 
the presented situation and to give a written statement they would use 
in the described situations. The presented situations are classified into 
two categories with regard to the relationship between the participants 
of the communication act: (1) situations involving interlocutors who are 
not in a familiar relationship (i.e., interlocutors who are not close and 
are not equal in terms of power relations, and communicate in more or 
less in/formal situations); (2) situations involving interlocutors in a fa-
miliar relationship (i.e., interlocutors in a close and equal relationship 
who communicate in more or less in/formal situations). Assignments of 
the two categories are organized into four pairs, asking respondents to 
share a speech act of similar propositional content: “I want you to return 
something that belongs to me” (for text of this role-playing task pair see 
Example 1 when interlocutors have (a) an unfamiliar relationship (label 
“NEFAM1”) and (b) a familiar relationship (label “FAM1”)); “I want you to 
answer my inquiry” (for text of this role-playing task pair see Example 2 
when interlocutors have (a) an unfamiliar relationship (label “NEFAM2”) 
198
Slovenščina 2.0, 2023 (1) | Articles
and (b) a familiar relationship (label “FAM2”)); “I want you to change 
something that bothers me” (for text of this role-playing task pair see 
Example 3 when interlocutors have (a) an unfamiliar relationship (label 
“NEFAM3”) and (b) a familiar relationship (label “FAM3”)); “I want you to 
stop behaving inappropriately” (for text of this role-playing task pair see 
Example 4 when interlocutors have (a) an unfamiliar relationship (label 
“NEFAM4”) and (b) a familiar relationship (label “FAM4”))10.
Example 1
(a)  Upravo si pojeo/la ručak u restoranu. Posluživao te stariji konobar koji 
se odnosio prema tebi ljubazno i profesionalno. Prilikom plaćanja 
računa konobar ti vraća 100 kuna manje nego što je trebao. Želiš 
da ti konobar vrati novac. Zamisli da se konobar nalazi pred tobom i 
napiši što bi mu točno rekao/la u danoj situaciji (nemoj prepričavati, 
već iskaz formuliraj kao da se izravno obraćaš sugovorniku).
 (Eng. You just ate lunch at a restaurant. You were served by an el-
derly waiter who treated you kindly and professionally. When pay-
ing the bill, the waiter refunds you 100 kunas less than he should 
have. You want the waiter to give you your money back. Imagine 
the waiter was in front of you and write what exactly you would 
say to him in the given situation (do not recount but formulate the 
statement as if you were addressing the interlocutor directly).)
(b)  Posudio/la si knjigu najboljem prijatelju (ili prijateljici). Rekao ti 
je da će ti je uskoro vratiti, no nije održao riječ. Sjedite zajedno u 
kafiću, situacija je opuštena, razgovarate o svakodnevnim stva-
rima. Želiš mu dati do znanja da ti treba čim prije vratiti knjigu. 
Zamisli da se tvoj prijatelj nalazi pred tobom i napiši što bi mu 
točno rekao/la u danoj situaciji (nemoj prepričavati, već iskaz for-
muliraj kao da se izravno obraćaš sugovorniku).
 (Eng. You lent a book to your best friend. (S)he told you (s)he’d give 
it back to you soon, but (s)he didn’t keep her/his word. You are 
sitting together in a café, the situation is relaxed, you talk about 
everyday things. You want to let her/him know you need to get your 
book back as soon as possible. Imagine your friend was in front of 
you and write what exactly you would say to her/him in the given 
situation (do not recount but formulate the statement as if you 
were addressing the interlocutor directly).)
10 Full texts of role-playing tasks are available in the corpus header as well.
199
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
Example 2
(a)  Poslao/la si e-mail profesoru s upitom možeš li pohađati njegov 
izborni kolegij i hitno trebaš njegov odgovor i potvrdu u mailu. 
Međutim, profesor ne odgovara već tjedan dana, a rok za upis 
završava sutradan. Želiš ponovno zatražiti njegovu povratnu in-
formaciju. Napiši kratak e-mail profesoru kakav bi mu uputio/la u 
navedenoj situaciji.
 (Eng. You sent an email to the professor asking if you can attend 
his elective course and urgently need his response and confirma-
tion in the email. However, the professor has not responded for a 
week, and the admission deadline ends the next day. You want to 
ask for his feedback again. Write a short email to the professor as 
you would in this situation.)
(b)  Poslao/la si poruku (WhatsApp, Viber, Messenger) najboljem pri-
jatelju (ili prijateljici) s pozivom na druženje sljedeće večeri. On je 
vidio poruku, ali nije odgovorio do sutradan. Želiš da ti odgovori 
čim prije kako bi mogao/la isplanirati ostatak dana. Napiši kratku 
poruku prijatelju kakvu bi mu uputio/la u navedenoj situaciji.
 (Eng. You sent a message (WhatsApp, Viber, Messenger) to your 
best friend with an invitation to hang out the next night. (S)he saw 
the message but did not respond until the next day. You want her/
him to reply as soon as possible so you can plan the rest of the day. 
Write a short message to your friend as you would in this situation.)
Example 3
(a)  Voziš se u taksiju. Prozori su otvoreni i želiš da ih taksist zatvori jer 
ti je hladno. Zamisli da se nalaziš u navedenoj situaciji i napiši što 
bi točno rekao/la taksistu (nemoj prepričavati, već iskaz formuliraj 
kao da se izravno obraćaš sugovorniku).
 (Eng. You’re riding in a cab. The windows are open, and you want 
the taxi driver to close them because you’re cold. Imagine that you 
are in this situation and write down what exactly you would say to 
the taxi driver (do not recount but formulate the statement as if you 
were addressing the interlocutor directly).)
(b)  Voziš se u autu na suvozačkom mjestu. Vozač je tvoj najbolji pri-
jatelj (ili prijateljica). Budući da vozi prebrzo i gleda u mobitel, ne 
osjećaš se ugodno i želiš da uspori. Zamisli da se nalaziš u danoj 
situaciji i napiši što bi mu točno rekao/la (nemoj prepričavati, već 
iskaz formuliraj kao da se izravno obraćaš sugovorniku).
200
Slovenščina 2.0, 2023 (1) | Articles
 (Eng. You are riding in the car in the passenger seat. The driver is 
your best friend. Because (s)he’s driving too fast and looking at her/
his cell phone, you don’t feel comfortable and want her/him to slow 
down. Imagine that you are in a given situation and write what 
exactly you would say to her/him (do not recount but formulate the 
statement as if you were addressing the interlocutor directly).)
Example 4
(a)  Nalaziš se u dućanu i čekaš u redu pred blagajnom. Velika je 
gužva. Ispred tebe se u red ugura gospođa srednje dobi. Ljudi u 
redu iza tebe negoduju jednako kao i ti. Želiš da gospođa stane 
na kraj reda. Zamisli da se nalaziš u danoj situaciji i napiši što bi 
točno rekao/la gospođi (nemoj prepričavati, već iskaz formuliraj 
kao da se izravno obraćaš sugovorniku).
 (Eng. You’re in the store waiting in line at the cash register. It’s 
crowded. A middle-aged lady squeezes in front of you. The people 
in line behind you are just as resentful as you are. You want the 
lady to stand at the end of the line. Imagine that you are in a given 
situation and write down what exactly you would say to the lady 
(do not recount but formulate the statement as if you were directly 
addressing the interlocutor).)
(b)  Slušaš predavanje na fakultetu. Sjediš pored dvoje kolega s koji-
ma si inače vrlo blizak/bliska. U jednom trenutku oni počinju glas-
no razgovarati i smijati se. Njihov razgovor ti smeta jer ne možeš 
pratiti predavanje, a i nastavnik pogledava u vašem smjeru. Želiš 
da prestanu. Zamisli da se nalaziš u danoj situaciji i napiši što bi 
im točno rekao/la (nemoj prepričavati, već iskaz formuliraj kao da 
se izravno obraćaš sugovorniku).
 (Eng. You’re listening to a lecture in college. You’re sitting next to 
two colleagues with whom you are otherwise very close. At some 
point, they start talking loudly and laughing. Their conversation 
bothers you because you can’t follow the lecture, and the lecturer 
looks in your direction. You want them to stop. Imagine that you 
are in a given situation and write down exactly what you would say 
to them (do not recount but formulate the statement as if you were 
addressing the interlocutor directly).)
The respondents were 100 Croatian speakers, all undergraduate 
(63%) or graduate students (37%) of the Faculty of Humanities and 
201
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
Social Sciences (University of Zagreb), aged between 18 to 33, with 
Croatian being the native language for the majority (96%). The ques-
tionnaire was administered in December 2020 and January 2021. Be-
fore completing the questionnaire, all the respondents were informed 
of the purposes of the study as well as what data would be collected. 
All the respondents voluntarily participated in the study and were made 
aware that they could withdraw from it at any time. By choosing to par-
ticipate in the study, respondents gave informed consent for their data 
to be processed for the stated research purposes. The questionnaire 
was administered anonymously via an online survey, and the language 
material collected was used exclusively for research purposes.
The elicitation of language production by the role-playing method 
has its advantages and disadvantages. On the one hand, it enables the 
collection of a large number of speech acts with the same proposition-
al content and illocutionary purpose. On the other hand, users of the 
corpus should keep in mind that the language material collected by 
this method does not reflect the features of actual language use, but 
instead shows what speakers think they would say and/or do in hypo-
thetical situations.
DirKorp contains 12,676 tokens and 1,692 types.11 Since it con-
sists of 800 speech acts, it is a relatively small corpus compared to 
some of the corpora with pragmatic annotation presented in Section 3. 
However, as the first Croatian corpus with detailed pragmatic annota-
tion, DirKorp can serve as a useful resource for researching the charac-
teristics of speech acts on a formal and content level, the application 
of politeness strategies in communication in different situations, and 
the properties of other grammatical-pragmatic and lexical-pragmatic 
phenomena in the Croatian language that are annotated in the corpus. 
In addition, we believe that DirKorp can serve as a complement to re-
search on speech acts that are conducted on authentic language mate-
rials and as a starting point for conducting contrastive research on the 
characteristics and use of speech acts in other languages. In addition, 
we hope that it will contribute to the development of larger corpora of 
11 Respondents’ answers contain utterances but also text about what they would do in the giv-
en situation. At this moment, the corpus contains no annotation of utterances of speech acts, 
and therefore we cannot analyse the average length of a response. Generally, we can only 
state that some speech acts contain only one utterance, while some contain more than one.
202
Slovenščina 2.0, 2023 (1) | Articles
the Croatian language with pragmatic annotation, and that such work 
will encourage a wider application of the corpus-pragmatic research 
method.
In Karlić and Bago (2021), we have conducted corpus pragmatic 
analyses of the collected speech acts to investigate ways and means 
of expressing directives, and their pragmatic characteristics and func-
tions. For example, we confirmed that indirect directives are more fre-
quent than direct ones, especially among interlocutors who are not in a 
familiar relationship. Regarding a(n) (un)familiar relationship between 
interlocutors, we detected that explicit illocutionary force is more fre-
quent in communication between interlocutors with a familiar relation-
ship, while implicit illocutionary force is more frequent in communica-
tion between interlocutors with an unfamiliar relationship. Additionally, 
we have identified that imperative utterances are a more frequent type 
of direct directives than utterances with a directive performative verb 
in 1st person. For more such corpus pragmatic analyses see Karlić and 
Bago (2021).
5 Corpus annotation
The collected language material has been manually annotated on 
the speech act level by two independent annotators12 with university 
graduate degrees in the field of philology. Annotators received oral and 
written instructions, including illustrative examples for all the features 
they had to annotate.
Basic categorization of speech acts (directive; direct and indirect; 
explicit and implicit) and their formal and pragmatic properties (i.e., 
performative verbs) was carried out according to the theory of speech 
acts by Austin (1962), Searle (1969; 1976), and their successors. The 
features and components of speech acts related to the phenomenon 
12 When comparing the annotations of two annotators, in all categories, disagreements were 
found in at most 1.2% of examples and were mostly the result of accidental mistakes by 
one of the annotators. Once such mistakes were corrected, a consensus was reached 
among the annotators. The only category in which the disagreement was higher (2.5%) 
was the category “Illocutionary force”. In most cases, these were examples with general-
ized conversational implicature (one annotator marked speech acts with this type of impli-
cature as explicit, and another as implicit). Based on the instruction to label speech acts 
with all types of implicature as implicit, a consensus was reached among the annotators for 
this category as well.
203
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
of politeness (familiarity, use of T/V forms, and certain lexical modifi-
ers of the illocutionary force of speech acts) are taken according to the 
politeness theory of Brown and Levinson (1987), while the grammati-
cal characteristics (utterance type, grammatical mood, modal verbs) of 
speech acts are categorized according to the grammatical descriptions 
of contemporary Croatian and Serbian languages (Silić and Pranjković, 
2007; Piper et al., 2005). For more on individual categories, see Karlić 
and Bago (2021). In the new version of DirKorp (v3.0), each speech act 
can contain up to 14 features. The first eight features were part of the 
corpus version v1.0, features nine to 12 were part of v2.0, while fea-
tures 13 and 14 were newly added. Appendix A contains the frequency 
distribution of features two to 14. For a more detailed frequency distri-
bution of all features see Karlić and Bago (2021).
(1)  Respondent ID – This mandatory feature contains information on 
the identification of the respondent uttering the speech act.
(2)  Familiarity/unfamiliarity – This mandatory feature contains in-
formation on the category of the proposed situation in which the 
speech act was uttered. Four situations are labelled ‘unfamiliar’ 
(involving interlocutors who are not in a familiar relationship), while 
the other four situations are labelled ‘familiar’ (involving interlocu-
tors who are in a familiar relationship).
(3)  Utterance type – This mandatory feature contains information 
on the utterance type regarding its structural organization. It con-
tains six labels: (a) an imperative utterance, (b) an assertive ut-
terance (a statement), (c) an utterance in the form of a question, 
(d) an utterance in the form of a predicate ellipsis13, (e) a nonver-
bal signal, (f) a case of avoidance of executing a speech act (see 
Example 5).
13 Utterances in the form of a predicate ellipsis were singled out as a separate category due to: 
(1) the absence of a verb (and potentially other components of the sentence structure) and 
therefore the default indirectness and implicitness of the speech act, which makes them 
incomparable to other utterances in the corpus; (2) impossibility to determine the type of 
utterance for all examples due to their elliptical structure.
204
Slovenščina 2.0, 2023 (1) | Articles
Example 5
(a)  E vrati mi onu knjigu koju sam ti posudio.
 (Eng. Hey, give me back that book I lent you.)
(b)  Oprostite, ali mislim da ste mi krivo vratili novce.
 (Eng. Excuse me, but I think you gave me my money back wrong.)
(c)  Možete li molim vas zatvoriti prozore?
 (Eng. Could you please close the windows?)
(d)  E, moja knjiga??
 (Eng. Hey, my book??)
(e)  [Samo bih zavrtjela očima da vide moje neodobravanje, ali ne bih 
ništa rekla.]14
 (Eng. [I’d just roll my eyes so that they see my disapproval, but I 
wouldn’t say anything.])
(f)  [Ne bih ništa rekao.]
 (Eng. [I wouldn’t say anything.])
(4)  Directive performative verb in 1st person – This optional feature 
contains information on the representation of a directive performa-
tive verb in 1st person as part of the speech act, only for assertive 
utterances and utterances in the form of a question. It contains two 
labels: (a) yes and (b) no (see Example 6).
Example 6
(a)  Oprostite, molim da odete na kraj reda.
 (Eng. Excuse me, I am imploring you to go to the end of the line.)
(b)  Gospođo, morate na kraj reda stati.
 (Eng. Madam, you must move to the end of the line.)
(5)  Illocutionary force – The optional feature contains information on 
the explicitness or implicitness of the illocutionary force of a speech 
act. It is only applied to utterances that contain verbal means (an 
imperative utterance, an assertive utterance, an utterance in the 
form of a question, and in the form of an ellipsis). It contains two 
labels: (a) explicit and (b) implicit (see Example 7).
14 Descriptions of non-verbal situations can be found in Example 5 (e) and (f). All other ex-
amples contain actual utterances. DirKorp v3.0 does not contain annotations of utterances. 
Therefore, it is currently not possible to filter the speech acts with regard to actual utterances 
or descriptions of non-verbal situations.
205
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
Example 7
(a)  Daj mi donesi više onu knjigu, treba mi!
 (Eng. Bring me that book already, I need it!)
(b)  Kaj je s onom knjigom koju sam ti posudio?
 (Eng. What happened to that book I lent you?)
(6)  Propositional content – This optional feature contains informa-
tion on the explicitness or implicitness of the propositional content 
of a speech act. It is only applied to utterances that contain verbal 
means (an imperative utterance, an assertive utterance, an utter-
ance in the form of a question, and in the form of an ellipsis). It 
contains two labels: (a) explicit and (b) implicit (see Example 8).
Example 8
(a)  Gledaj na cestu, pusti mobitel.
 (Eng. Look at the road, leave the cell phone.)
(b)  Ti hoćeš da poginemo?
 (Eng. You want us to die?)
(7)  T/V form – This optional feature contains information on how the 
respondent addressed the interlocutor, using an informal (T-form) 
or a formal you (V-form). It is only applied to utterances that con-
tain verbal means (an imperative utterance, an assertive utterance, 
an utterance in the form of a question, and in the form of an ellip-
sis). It contains three labels: (a) T-form, (b) V-form, and (c) impos-
sible to determine (see Example 9).
Example 9
(a)  Oprosti, dao si mi manje novca
 (Eng. SorryT-form, youT-form gave me less change.)
(b)  Oprostite, mislim da ste mi ipak još dužni 100 kuna.
 (Eng. ExcuseV-form me, I think youV-form still owe me 100 kunas.)
(c)  Hmm... još 100 kuna, zar ne?
 (Eng. Hmm… another 100 kunas, right?)
(8)  Exhortative – This optional feature contains information on the 
representation of an exhortative as part of the speech act (a lexical 
206
Slovenščina 2.0, 2023 (1) | Articles
mean used to express encouragement, i.e., incentive particles). It 
is only applied to utterances that contain verbal means (an impera-
tive utterance, an assertive utterance, an utterance in the form of a 
question, and in the form of an ellipsis). It contains two labels: (a) 
yes and (b) no (see Example 10).
Example 10
(a)  Daj mi više vrati knjigu, treba mi za knjižnicu.
 (Eng. Bring me back my book already, I need it for the library.)
(b)  Jel se sjećaš one knjige koju sam ti posudila? Potrebna mi je. 
Možeš li mi ju donijeti sutra na faks?
 (Eng. Do you remember that book I lent you? I need it. Could you 
bring it tomorrow to uni?)
(9)  Request – This optional feature contains information on whether 
the speech act includes a lexical marker of request (e.g., “please”). 
It is only applied to utterances that contain verbal means (an im-
perative utterance, an assertive utterance, an utterance in the form 
of a question, and in the form of an ellipsis). It contains two labels: 
(a) yes and (b) no (see Example 11).
Example 11
(a) E da, jel bi mi mogao/la vratiti knjigu, molim te?
(Eng. Oh yeah, could you bring the book back, please?)
(b) Zaboravio si mi vratiti knjigu, jel se možeš idući put sjetiti?
(Eng. You forgot to bring me back the book, can you remember next time?)
(10) Apology – This optional feature contains information on whether 
the speech act includes a lexical marker of apology. It is only ap-
plied to utterances that contain verbal means (an imperative ut-
terance, an assertive utterance, an utterance in the form of a ques-
tion, and in the form of an ellipsis). It contains two labels: (a) yes 
and (b) no (see Example 12).
 
Example 12
(a)  Oprostite, ovdje fali još 100 kuna
 (Eng. Excuse me, 100 kunas are missing here.)
207
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
(b)  Možete li molim vas pritvoriti prozore, hladno mi je?
 (Eng. Could you please close the windows, I’m cold?)
(11) Gratitude – This optional feature contains information on whether 
the speech act includes a lexical marker of gratitude. It is only ap-
plied to utterances that contain verbal means (an imperative ut-
terance, an assertive utterance, an utterance in the form of a ques-
tion, and in the form of an ellipsis). It contains two labels: (a) yes 
and (b) no (see Example 13).
 Example 13
(a) Molim te mi samo javi da znam zbog organizacije hoćeš li doći. 
Hvala ti!
 (Eng. Please just let me know whether you’re coming so that I 
know because of the organization. Thank you!)
(b) Heej, jel dolaziš večeras na druženje? Moram znati zbog organi-
zacije. xoxo
 (Eng. Heeey, are you coming tonight to hang out? I need to know 
because of the organization. xoxo)
(12) Honorific title – This optional feature contains information on 
whether the speech act includes an honorific title. It is only ap-
plied to utterances that contain verbal means (an imperative ut-
terance, an assertive utterance, an utterance in the form of a ques-
tion, and in the form of an ellipsis). It contains two labels: (a) yes 
and (b) no (see Example 14).
 Example 14
(a)  Gospođo, kraj reda je dolje.
 (Eng. Madam, the end of the line is back there.)
(b)  Oprostite, tamo je kraj reda!
 (Eng. Excuse me, the end of the line is there!)
(13) Grammatical mood – This optional feature contains information 
on grammatical mood used in a speech act. It is only applied to 
indirect speech acts (assertive utterances and utterances in the 
form of a question) since it is understood that direct imperative 
208
Slovenščina 2.0, 2023 (1) | Articles
speech acts contain verbs in the imperative mood. Accordingly, 
this feature contains two labels: (a) indicative mood and (b) condi-
tional mood (see Example 15).
Example 15
(a)  Oprostite, ali ovo nije kraj reda.
 (Eng. Excuse me, but this is not the end of the line.)
(b)  Oprostite jel bi mogli zatvorit prozore? Malo mi je hladno.
 (Eng. Excuse me, could you close the windows? I’m a little cold.)
(14) Modal verb in 2nd person – This optional feature contains infor-
mation on the representation of modal verb in 2nd person as part of 
a speech act. It is only applied to indirect speech acts (an assertive 
utterance and an utterance in the form of a question). It contains 
two labels: (a) yes and (b) no (see Example 16).
Example 16
(a)  Oprostite, mislim da je došlo do pogreške, trebate mi vratiti još 
100 kuna.
 (Eng. Sorry, I think there was a mistake, you have to return an-
other 100 kunas.)
(b)  Malo je hladno ovdje, možemo možda zatvoriti prozor?
 (Eng. It’s a little cold in here, can we possibly close the window?)
6 Corpus format
DirKorp is encoded according to the TEI P5: Guidelines for Electronic 
Text Encoding and Interchange, developed and maintained by the Text 
Encoding Initiative Consortium (TEI) (TEI Consortium, 2021). The TEI 
document is comprised of a header and the body of the corpus. The 
content of the elements and attributes are in Croatian. The metadata 
of the corpus is given in the header including bibliographic informa-
tion; the editorial practice; a structured taxonomy describing catego-
ries used for each of the 14 pragmatic features in the annotation pro-
cess (see Figure 1 for an example), including the full text of the eight 
situations on the questionnaire; a list of questionnaire participants with 
information on their age, gender, undergraduate or graduate level of 
209
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
study, enrolment in a philological/non-philological/combined study 
program and native language (see Figure 2 for an example); and a list 
of revisions of the DirKorp versions. The body of the corpus is com-
posed of one division containing utterances with pragmatic features 
(see Figure 3 for an example).
DirKorp is available for download under the CC BY-SA 4.0 license 
from GitHub in TEI format (https://github.com/pbago/DirKorp).
<taxonomy xml:id=”tiVi”>
<category xml:id=”ti”>
<catDesc>Govorni čin sadržava obraćanje na ti (atribut se 
odnosi na tipove iskaza koji uključuju verbalna sredstva 
[imperativni, tvrdnja, upitni, eliptični]).</catDesc>
</category>
<category xml:id=”vi”>
<catDesc>Govorni čin sadržava obraćanje na Vi (atribut se 
odnosi na tipove iskaza koji uključuju verbalna sredstva 
[imperativni, tvrdnja, upitni, eliptični]).</catDesc>
</category>
<category xml:id=”persNeodredivo”>
<catDesc>Nije moguće odrediti sadržava li govorni čin 
obraćanje na ti ili Vi (atribut se odnosi na tipove iskaza 
koji uključuju verbalna sredstva [imperativni, tvrdnja, 
upitni, eliptični]).</catDesc>
</category>
</taxonomy>
Figure 1: An example of a pragmatic feature description – how the respondent ad-
dressed the interlocutor (V-form, T-form, or impossible to determine, annotation feature 
7 from Section 5).
<person xml:id=”I001” sex=”F”>
<p>ispitanik/ispitanica, 20 godina, spol Ž, preddiplomski 
studij Filozofskog fakulteta, nefilološko usmjerenje, materinji 
jezik hrvatski</p>
</person>
Figure 2: An example of participant information.
<u who=”#I001” ana=”#NEFAM1 #tvrdnja #dpg1N #isI #psI #vi 
#adhorativN #molbaN #isprikaY #zahvalaN #honorifikN #gnI 
#mg2N”>Ispričavam se, pardon, fali još sto kuna. Oprostite.</u>
Figure 3: An example of a speech act containing all 14 pragmatic features.
210
Slovenščina 2.0, 2023 (1) | Articles
7 Conclusion and future work
In this article we have presented DirKorp v3.0, the first Croatian cor-
pus of directive speech acts, containing 800 elicited speech acts col-
lected via an online questionnaire with role-playing tasks, specifically 
developed for pragmatic research studies. The respondents were 100 
Croatian speakers, all students of the Faculty of Humanities and Social 
Science (University of Zagreb). The corpus has been manually anno-
tated on the level of a speech act, with each speech act containing up 
to 14 features. It contains 12,676 tokens and 1,692 types. The corpus 
is available for download under the CC BY-SA 4.0 license from GitHub 
in TEI format.
Further work is planned on the corpus, which includes an evalu-
ation of the developed schema for annotating directive speech acts 
(e.g., test-retest reliability on a sample of data to evaluate stability and 
consistency of the schema, domain experts reviewing the schema to 
determine if it adequately captures the relevant aspects of the data, 
reviewing the adequacy of encoding choices regarding attributes and 
its values), annotation at the levels smaller than a speech act, as well 
as augmentation with additional features such as information on vari-
ous politeness strategies applied in a speech act.
Acknowledgments
This paper is generously co-financed by the institutional project of the 
Faculty of Humanities and Social Sciences (University of Zagreb) “South 
Slavic languages in use: pragmatic analyses” (principle researcher Vir-
na Karlić). We also wish to thank our annotators for the time and effort.
References
Allen, J. F., Schubert, L. K., Ferguson, G., Heeman, P., Hwang, C. H., Kato, T., 
Light, M., …, & Traum, D. R. (1995). The TRAINS Project: A Case Study 
in Building a Conversational Planning Agent. Journal of Experimental & 
Theoretical Artificial Intelligence, 7(1),7–48.
Alsop, S., & Nesi, H. (2013). Annotating a Corpus of Spoken English: The Engi-
neering Lecture Corpus (ELC). In Proceedings of GSCP 2012: Speech and 
Corpora (pp. 58–62). Firenze University Press, Florence.
211
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
Alsop, S., & Nesi, H. (2014). The Pragmatic Annotation of a Corpus of Aca-
demic Lectures. In The International Conference on Language Resources 
and Evaluation 2014 Proceedings (pp. 1560–1563). Reykjavik: European 
Language Resources Association.
Anderson, A. H., Bader, M., Gurman Bard, E., Boyle, E., Doherty, G., Garrod, S., 
Isard, S., …, & Weinert, R. (1991). The HCRC Map Task Corpus, Language 
and Speech, 34(4), 351–366.
Austin, J. L. (1962). How to Do Things with Words. Oxford: Clarendon Press. 
Barron, A. (2008). The structure of requests in Irish English and English Eng-
lish. In K. P. Schneider & A. Barron (Eds.), Variational Pragmatics: A Focus 
on Regional Varieties in Pluricentric Languages (pp. 35–68). John Benja-
mins Publishing Company.
Brown, P., & Levinson, S. C. (1987). Politeness: Some Universals in Language 
Usage. Cambridge University Press.
Bunt, H. (2017). Computational Pragmatics. In Oxford Handbook of Pragmat-
ics (pp. 326–345). Oxford University Press, New York.
Bunt, H., Petukhova, V., Malchanau, A., Fang, A. & Wijnhoven, K. (2019). The 
DialogBank: Dialogues with Interoperable Annotations. In Language Re-
sources and Evaluation, 53(2), 213–249.
Capone, A. (2009). Speech Acts, Classification and Definition. In Concise En-
cyclopedia of Pragmatics (pp. 1015–1017). Oxford: Elsevier.
Caspers, J. (2000). Melodic Characteristics of Backchannels in Dutch Map Task 
Dialogues. In Proceedings, 6th International Conference on Spoken Lan-
guage Processing (pp. 611–614). Beijing: China Military Friendship Pub-
lish,. Retrieved from https://www.isca-speech.org/archive/icslp_2000/
Flӧck, I., & Geluykens, R. (2015). Speech Acts in Corpus Pragmatics: A Quan-
titative Contrastive Study of Directives in Spontaneous and Elicited Dis-
course. In Yearbook of Corpus Linguistics and Pragmatics (pp. 7–37). 
Springer International Publishing.
Franović, T., & Šnajder, J. (2012). Speech Act Based Classification of Email 
Messages in Croatian Language. In Proceedings of the Eighth Language 
Technologies Conference (pp. 69–72). Ljubljana: Information Society.
Geertzen, J., Girard, Y., Morante, R., Van der Sluis, J., Van Dam, H., Suijker-
buijk, B., Van der Werf, R., & Bunt, H. (2004). The DIAMOND Project. In: 
Proceedings of the 8th Workshop on the Semantics and Pragmatics of Dia-
logue (CATALOG 2004), Barcelona.
Godfrey, J., Holliman, E. & McDaniel, J. (1992). SWITCHBOARD: Telephone 
Speech Corpus for Research and Development. In: IEEE International 
212
Slovenščina 2.0, 2023 (1) | Articles
Conference on Acoustics, Speech, and Signal Processing (Vol. 1, pp. 517–
520). San Francisco: IEEE Computer Society.
Hržica, G., Košutar, S., & Posavec, K. (2021). Konektori i druge diskursne 
oznake u pisanome i spontanome govorenom jeziku. Fluminensia: časopis 
za filološka istraživanja, 33(1), 25–52.
Huang, Y. (2009). Speech Acts. In Concise Encyclopedia of Pragmatics (pp. 
1000–1009). Oxford: Elsevier.
Ivanetić, N. (1995). Govorni činovi. Zagreb: FF-press, Zavod za lingvistiku Filo-
zofskoga fakulteta Sveučilišta u Zagrebu.
Jucker, A. H. (2009). Speech Act Research between Armchair, Field and Labo-
ratory: The Case of Compliments. Journal of Pragmatics, 41, 1611–1635.
Jucker, A. H., Schreier, D., & Hundt, M. (Eds.). (2009). Corpora: Pragmatics and 
Discourse. Rodopi, Amsterdam.
Kallen, J. L., & Kirk, J. M. (2012). SPICE-Ireland: A User’s Guide. Retrieved from 
https://pure.qub.ac.uk/en/publications/spice-ireland-a-users-guide
Karlić, V., & Bago, P. (2021). (Računalna) pragmatika: temeljni pojmovi i 
korpusnopragmatičke analize. Zagreb: FF Press. Retrieved from https://
openbooks.ffzg.unizg.hr/index.php/Ffpress/catalog/book/125.
Kehoe, A., & Gee, M. (2007). New Corpora from the Web: Making Web Text 
More ‘Text-Like’. In Studies in Variation, Contacts and Change in English 2. 
Retrieved from https://varieng.helsinki.fi/series/volumes/02/kehoe_gee/
Kehoe, A., & Gee, M. (2012). Reader Comments as an Aboutness Indicator 
in Online Texts: Introducing the Birmingham Blog Corpus. In: Studies in 
Variation, Contacts and Change in English 12. Retrieved from https://var-
ieng.helsinki.fi/series/volumes/12/kehoe_gee/
Kuvač Kraljević, J., & Hržica, G. (2016). Croatian Adult Spoken Language Cor-
pus (HrAL). Fluminensia: časopis za filološka istraživanja, 28(2), 87–102.
Leech, G. N. (1992). Corpora and Theories of Linguistic Performance. In Direc-
tions in Corpus Linguistics (pp. 105–122). De Gruyter, Berlin.
Ljubešić, N., & Klubička, F. (2014). {bs, hr, sr}WaC-Web Corpora of Bosnian, 
Croatian and Serbian. In: Proceedings of the 9th Web as Corpus Workshop 
(WaC-9) (pp. 29–35). Association for Computational Linguistics, Gothen-
burg. Retrieved from https://aclanthology.org/W14-0405.pdf
Lutzky, U., & Kehoe, A. (2017a). I Apologize for My Poor Blogging: Searching for 
Apologies in the Birmingham Blog Corpus. Corpus Pragmatics, 1(1), 37–56.
Lutzky, U., & Kehoe, A. (2017b). Oops, I Didn’t Mean to Be so Flippant. A Cor-
pus Pragmatic Analysis of Apologies in Blog Data. Journal of Pragmatics, 
116, 27–36.
213
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
Matić, D. (2011). Govorni činovi u političkome diskursu. PhD thesis. Zagreb: 
Faculty of Humanities and Social Sciences.
Miščević, N. (2018). Rođenje pragmatike. Orion Art, Beograd.
Palašić, N. (2020). Pragmalingvistika – lingvistički pravac ili petlja? Zagreb: 
Hrvatska sveučilišna naklada.
Petukhova, V., Gropp, M., Klakow, D., Eigner, G., Topf, M., Srb, S., Motlicek, P., … 
Potard, …, & Schmidt, A. (2014). The DBOX Corpus Collection of Spoken 
Human-Human and Human-Machine Dialogues. In Proceedings of the Ninth 
International Conference on Language Resources and Evaluation (LREC’14) 
(pp. 252–258). European Language Resources Association, Reykjavik.
Piper, P. et al. (2005) = Предраг Пипер, Ивана Антонић, Бранислава Ружић, 
Срето Танасић, Људмила Поповић, Бранко Тошовић. 2005. Синтакса 
савременог српског језика. Проста реченица, Београд: Институт за 
српски језик САНУ, Београдска књига, Матица српска.
Pišković, T. (2007). Dramski diskurs između pragmalingvistike i feminističke 
lingvistike. Rasprave: Časopis Instituta za hrvatski jezik i jezikoslovlje, 
33(1), 325–341.
Prasad, R., Webber, B., & Lee, A. (2018). Discourse Annotation in the PDTB: 
The NextGeneration. In: Proceedings of the 14th Joint ACL-ISO Workshop 
on Interoperable Semantic Annotation (pp. 87–97). Santa Fe: Association 
for Computational Linguistics. Retrieved from https://aclanthology.org/
W18-4710.pdf
Prüst, H., Minnen, G. & Beun, R. (1984). Transcriptie dialooogesperiment juni/
juli 1984, IPORapport 481. Eindhoven: Institute for Perception Research, 
Eindhoven University of Technology. 
Pupovac, M. (1990). Jezik i djelovanje. Zagreb: Biblioteka časopisa Pitanja.
Romero-Trillo, J. (Ed.). (2008). Pragmatics and Corpus Linguistics: A Mutualis-
tic Entente. De Gruyter, Berlin.
Rühlemann, C., & Aijmer, K. (2015). Introduction. Corpus pragmatics: laying 
the foundations. In: Corpus pragmatics (pp. 1–28).
Searle, J. R. (1969). Speech Acts. Cambridge University Press, Cambridge.
Searle, J. R. (1975). A Taxonomy of Speech Acts. In: Minnesota Studies in 
the Philosophy of Science (Vol. 9, pp. 344–369). University of Minnesota 
Press.
Searle, J. R. (1976). A classification of illocutionary acts. Language in Society, 
5, 1–23.
Silić, S. & Pranjković, I. (2007). Gramatika hrvatskoga jezika za gimnazije i vi-
soka učilista. Zagreb: Školska knjiga.
214
Slovenščina 2.0, 2023 (1) | Articles
Šegić, T. (2019). Tata kupi mi auto und Nivea Milk weil es nichts Besseres für 
die Hautpflege gibt. Filologija, 73, 103–116.
Tadić, M. (1996). Računalna obradba hrvatskoga i nacionalni korpus. Suvre-
mena lingvistika, 41–42, 603–611.
TEI Consortium (Ed.). (2021). TEI P5: Guidelines for Electronic Text Encoding 
and Interchange. TEI Consortium.
Trosborg, A. (1995). Interlanguage Pragmatics: Requests, Complaints, and 
Apologies. Berlin; New York: Mouton de Gruyter.
Wojtaszek, A. (2016). Thirty years of Discourse Completion Test in Contrastive 
Pragmatics research, Linguistica Silesiana, 37, 161–173.
Yule, G. (2002). Pragmatics. Oxford, New York: Oxford University Press.
DirKorp: hrvaški korpus direktivnih govornih dejanj (v3.0)
V prispevku predstavljamo razvoj nove različice (v3.0) korpusa DirKorp (Kor-
pus direktivnih govornih činova hrvatskoga jezika), prvega hrvaškega korpusa 
direktivnih govornih dejanj, ki je bil izdelan za namene raziskav pragmatike. 
Korpus vsebuje 800 govornih dejanj, ki so bila zbrana s spletnim vprašalnikom 
z nalogami igranja vlog – gre za metodo stimulirane komunikacije, ki poteka 
pod vnaprej določenimi pogoji. Metoda je primerna za raziskovanje govornih 
dejanj, saj lahko na ta način zberemo veliko število primerov z enako propozi-
cijsko vsebino in ilokucijskim namenom, ki so uporabljeni v enaki kontrolira-
ni situaciji. Predstavljene situacije razdelimo v dve kategoriji glede na odnos 
med udeleženci komunikacijskega dejanja: (1) situacije, ki vključujejo sogo-
vorce, ki niso v sorodstvenem razmerju; (2) situacije z govorci v sorodstvenem 
razmerju. Naloge v obeh kategorijah so razdeljene v štiri pare, od sodelujočih 
pa zahtevajo, da pripišejo govorno dejanje s podobno propozicijsko vsebino. 
V vprašalniku je sodelovalo 100 govorcev hrvaščine; vsi so bili dodiplomski 
(63 %) ali podiplomski študenti (37 %) Fakultete za humanistiko in družbene 
vede (Univerza v Zagrebu). Korpus je bil ročno označen na ravni govornih de-
janj, vsako dejanje pa vsebuje do 14 značilnosti: (1) ID sodelujočega, (2) so-
rodstveno/nesorodstveno razmerje, (3) tip izjave, (4) direktivni performativni 
glagol v prvi osebi, (5) ilokucijska sila, (6) propozicijska vsebina, (7) tikanje/
vikanje, (8) prepričevalnost, (9) leksikalni označevalec za prošnjo, (10) leksi-
kalni označevalec za opravičilo, (12) naziv spoštovanja, (13) slovnični naklon, 
(14) modalni glagol v drugi osebi. Korpus vsebuje 12.676 pojavnic in 1.692 
različnic, enkodiran pa je v skladu s smernicami TEI P5: Guidelines for Elec-
tronic Text Encoding and Interchange, ki jih razvija in vzdržuje konzorcij Text 
215
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
Encoding Initiative Consortium (TEI). DirKorp je v formatu TEI na voljo za pre-
nos pod licenco CC BY-SA 4.0 na platformi GitHub. V prispevku opišemo ozna-
čevanje in strukturo korpusa.
Ključne besede: korpusna pragmatika, direktivna govorna dejanja, DirKorp, 
hrvaški jezik
216
Slovenščina 2.0, 2023 (1) | Articles
Appendix A: Frequency distribution of annotated features 2-14
 
Utterance type ∑ Illocutionary force Propositional content T/V form Exhortative Request Apology Gratitude Honorific title Grammatical mood Modal verb in 2nd person
Yes No Explicit Implicit Explicit Implicit T-form V-form Yes No Yes No Yes No Yes No Yes No Indicative Conditional Yes No
A
NEFAM1
Imperative 2 N/A N/A 2 0 0 2 0 2 0 1 1 2 0 1 1 0 2 0 2 N/A N/A N/A N/A
Assertive 88 3 85 3 85 11 77 2 86 0 0 88 3 85 88 0 0 88 4 84 83 5 11 77
Question 10 0 10 0 10 0 10 0 9 1 0 10 2 8 8 2 0 10 1 9 8 2 9 1
Ellipsis 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
B
FAM1
Imperative 22 N/A N/A 22 0 21 1 22 0 0 17 5 7 15 0 22 0 22 0 22 N/A N/A N/A N/A
Assertive 15 1 14 1 14 7 8 15 0 0 1 14 1 14 0 15 0 15 0 15 8 7 2 13
Question 60 0 60 0 60 33 27 60 0 0 2 58 7 53 3 57 0 60 0 60 55 5 28 32
Ellipsis 3 N/A N/A 0 3 2 1 2 0 1 0 3 0 3 0 3 0 3 0 3 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
C
NEFAM2
Imperative 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Assertive 87 39 48 39 48 56 31 0 87 0 0 87 44 43 10 77 40 47 66 21 54 33 1 86
Question 13 0 13 0 13 12 1 0 87 0 0 13 6 7 2 11 6 7 10 3 12 1 11 2
Ellipsis 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
D
FAM2
Imperative 40 N/A N/A 40 0 38 2 40 0 0 14 26 13 27 0 40 2 38 0 40 N/A N/A N/A N/A
Assertive 8 2 6 2 6 4 4 8 0 0 1 7 2 6 0 8 1 8 0 8 6 2 1 7
Question 46 0 45 0 46 5 41 45 0 1 0 46 2 44 0 46 0 46 0 46 45 1 13 33
Ellipsis 3 N/A N/A 0 3 1 2 1 0 2 0 3 0 3 1 2 1 2 0 3 N/A N/A N/A N/A
Nonverbal signal 2 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 1 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
E
NEFAM3
Imperative 2 N/A N/A 2 0 2 0 0 2 0 0 2 0 2 1 1 1 1 0 1 N/A N/A N/A N/A
Assertive 1 0 1 0 1 1 0 0 1 0 0 1 0 1 0 1 1 0 0 1 1 0 0 1
Question 96 1 95 1 95 96 0 0 95 1 0 96 42 54 63 33 4 92 3 93 71 25 84 12
Ellipsis 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 1 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
F
FAM3
Imperative 86 N/A N/A 86 0 80 6 86 0 0 59 27 16 70 0 86 0 86 0 86 N/A N/A N/A N/A
Assertive 4 0 4 0 4 1 3 4 0 0 1 3 0 4 0 4 0 4 0 4 2 2 0 4
Question 9 0 9 0 9 5 4 9 0 0 0 9 2 7 0 9 0 9 0 9 8 1 7 2
Ellipsis 1 N/A N/A 0 1 0 1 1 0 0 1 0 1 0 0 1 0 1 0 1 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
H
NEFAM4
Imperative 19 N/A N/A 19 0 19 0 0 19 0 3 16 11 8 4 15 0 19 14 5 N/A N/A N/A N/A
Assertive 55 9 46 9 46 12 43 0 55 0 0 55 10 45 27 28 0 55 35 20 52 3 4 51
Question 12 0 12 0 12 10 2 0 11 1 0 12 3 9 7 5 0 12 5 7 11 1 9 3
Ellipsis 1 N/A N/A 0 1 1 0 0 1 0 0 1 1 0 0 1 0 1 1 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 13 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
G
FAM4
Imperative 43 N/A N/A 43 0 40 3 1 0 42 27 16 11 32 1 42 0 43 0 43 N/A N/A N/A N/A
Assertive 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 1 0 1 1 0 0 1
Question 12 0 12 0 12 12 0 0 0 12 1 11 6 6 1 11 1 11 0 12 12 0 12 0
Ellipsis 37 N/A N/A 0 37 33 4 1 1 35 22 15 2 35 0 37 0 37 0 37 N/A N/A N/A N/A
Nonverbal signal 2 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 5 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Directive performative 
verb in 1st person 
to 
determine
217
DirKorp: A Croatian Corpus of Directive Speech Acts (v3.0)
 
Utterance type ∑ Illocutionary force Propositional content T/V form Exhortative Request Apology Gratitude Honorific title Grammatical mood Modal verb in 2nd person
Yes No Explicit Implicit Explicit Implicit T-form V-form Yes No Yes No Yes No Yes No Yes No Indicative Conditional Yes No
A
NEFAM1
Imperative 2 N/A N/A 2 0 0 2 0 2 0 1 1 2 0 1 1 0 2 0 2 N/A N/A N/A N/A
Assertive 88 3 85 3 85 11 77 2 86 0 0 88 3 85 88 0 0 88 4 84 83 5 11 77
Question 10 0 10 0 10 0 10 0 9 1 0 10 2 8 8 2 0 10 1 9 8 2 9 1
Ellipsis 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
B
FAM1
Imperative 22 N/A N/A 22 0 21 1 22 0 0 17 5 7 15 0 22 0 22 0 22 N/A N/A N/A N/A
Assertive 15 1 14 1 14 7 8 15 0 0 1 14 1 14 0 15 0 15 0 15 8 7 2 13
Question 60 0 60 0 60 33 27 60 0 0 2 58 7 53 3 57 0 60 0 60 55 5 28 32
Ellipsis 3 N/A N/A 0 3 2 1 2 0 1 0 3 0 3 0 3 0 3 0 3 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
C
NEFAM2
Imperative 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Assertive 87 39 48 39 48 56 31 0 87 0 0 87 44 43 10 77 40 47 66 21 54 33 1 86
Question 13 0 13 0 13 12 1 0 87 0 0 13 6 7 2 11 6 7 10 3 12 1 11 2
Ellipsis 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
D
FAM2
Imperative 40 N/A N/A 40 0 38 2 40 0 0 14 26 13 27 0 40 2 38 0 40 N/A N/A N/A N/A
Assertive 8 2 6 2 6 4 4 8 0 0 1 7 2 6 0 8 1 8 0 8 6 2 1 7
Question 46 0 45 0 46 5 41 45 0 1 0 46 2 44 0 46 0 46 0 46 45 1 13 33
Ellipsis 3 N/A N/A 0 3 1 2 1 0 2 0 3 0 3 1 2 1 2 0 3 N/A N/A N/A N/A
Nonverbal signal 2 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 1 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
E
NEFAM3
Imperative 2 N/A N/A 2 0 2 0 0 2 0 0 2 0 2 1 1 1 1 0 1 N/A N/A N/A N/A
Assertive 1 0 1 0 1 1 0 0 1 0 0 1 0 1 0 1 1 0 0 1 1 0 0 1
Question 96 1 95 1 95 96 0 0 95 1 0 96 42 54 63 33 4 92 3 93 71 25 84 12
Ellipsis 0 N/A N/A 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 1 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
F
FAM3
Imperative 86 N/A N/A 86 0 80 6 86 0 0 59 27 16 70 0 86 0 86 0 86 N/A N/A N/A N/A
Assertive 4 0 4 0 4 1 3 4 0 0 1 3 0 4 0 4 0 4 0 4 2 2 0 4
Question 9 0 9 0 9 5 4 9 0 0 0 9 2 7 0 9 0 9 0 9 8 1 7 2
Ellipsis 1 N/A N/A 0 1 0 1 1 0 0 1 0 1 0 0 1 0 1 0 1 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
H
NEFAM4
Imperative 19 N/A N/A 19 0 19 0 0 19 0 3 16 11 8 4 15 0 19 14 5 N/A N/A N/A N/A
Assertive 55 9 46 9 46 12 43 0 55 0 0 55 10 45 27 28 0 55 35 20 52 3 4 51
Question 12 0 12 0 12 10 2 0 11 1 0 12 3 9 7 5 0 12 5 7 11 1 9 3
Ellipsis 1 N/A N/A 0 1 1 0 0 1 0 0 1 1 0 0 1 0 1 1 0 N/A N/A N/A N/A
Nonverbal signal 0 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 13 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
G
FAM4
Imperative 43 N/A N/A 43 0 40 3 1 0 42 27 16 11 32 1 42 0 43 0 43 N/A N/A N/A N/A
Assertive 1 0 1 0 1 0 1 0 0 1 0 1 0 1 0 1 0 1 0 1 1 0 0 1
Question 12 0 12 0 12 12 0 0 0 12 1 11 6 6 1 11 1 11 0 12 12 0 12 0
Ellipsis 37 N/A N/A 0 37 33 4 1 1 35 22 15 2 35 0 37 0 37 0 37 N/A N/A N/A N/A
Nonverbal signal 2 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Avoidance 5 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
Directive performative 
verb in 1st person 
to 
determine