353
Aatu Liimatta DOI: 10.4312/linguistica.63.1-2.353-374
aatu.liimatta@helsinki.fi
Jani Marjanen
jani.marjanen@helsinki.fi
Tuuli Tahko
tuuli.tahko@alumni.helsinki.fi
Mikko Tolonen
mikko.tolonen@helsinki.fi
Tanja Säily
tanja.saily@helsinki.fi
University of Helsinki
DIMENSIONS OF INCOMING ECONOMIC VOCABULARY IN 
EIGHTEENTH-CENTURY BRITAIN
1 INTRODUCTION
The intellectual history of the eighteenth century, characterized by the emergence of 
the Enlightenment and its platform of modernity, has been increasingly linked to eco-
nomic improvement in contemporary discourse (see especially Robertson 2005). Some 
scholars also use descriptions such as “economic enlightenment” (Popplow 2010) or 
“agricultural enlightenment” (Jones 2016) to foreground the whole period as particu-
larly economic or agricultural. The crux lies not in a conventional definition of Enlight-
enment’s nature, but rather in comprehending the practical evolution of eighteenth-cen-
tury thought. This interest in the economy is reflected in changes in eighteenth-century 
vocabulary, which is often pointed out in scholarship (Shovlin 2006: 2–5, McIntosh 
2020: 163–165), but not studied rigorously in terms of language use, and only men-
tioned briefly in e.g. overviews of English historical lexis (Nevalainen 1999). Bi9ber/
Finegan (1997) have shown that from the eighteenth century onwards, English registers 
diverged in their language use: speech-based and popular written registers like fic-
tion became more colloquial, whereas expository ‘specialized’ registers like academic 
texts became more literate in style.1 As the English language underwent significant 
functional expansion stimulated by various socio-cultural changes during the eighteenth 
1 “Literate” in this context is to be understood as opposed to “oral” or “conversational”; according to Biber 
& Finegan (1997: 260), literate registers are “characterized by careful, sustained production circumstances; 
informational communicative purposes; and minimal interactiveness”.
354
century (Culpeper/Nevala 2012), and considering the contemporary interest in the 
economy, we hypothesize that this expansion drove the spread of new economic vo-
cabulary both in terms of its frequency of use but also in its application in both con-
crete and increasingly abstract contexts. The main emphasis of our paper is to utilise a 
data-driven approach to draw conclusions about economic vocabulary. While we have 
identified various potential factors that may have influenced its expansion, our goal in 
this study is to primarily observe and analyze the data, rather than getting involved in 
a discussion about the underlying reasons. This is particularly the case with regard to 
potential underlying economic developments that may or may not have influenced the 
linguistic expressions. While they are interesting, the correlations are difficult to chart 
due to the difficulty in comparing linguistic and economic data. 
In combining linguistic and historical expertise, we make use of novel compu-
tational methods and draw on the massive database of Eighteenth Century Collec-
tions Online (ECCO) to trace incoming economic vocabulary in British publications 
dealing with economic matters in the widest sense (about ECCO, see Tolonen et 
al. 2022). As the starting point of our analysis of economic vocabulary, we focus 
on the words classified by the Historical Thesaurus of the Oxford English Diction-
ary under the category ‘trade and finance’. In order to identify incoming economic 
vocabulary in particular, we apply a novel method, the Proportional Change Rate 
(PCR; Nevalainen et al. 2020), a measure developed for identifying periods of rapid 
and slower linguistic change as well as changing linguistic items. Furthermore, in 
order to gain more insight into the lexical functions and underlying discourses of the 
incoming vocabulary, we make use of an approach inspired by the multi-dimensional 
method of register analysis (MDA; Biber 1988) in which we apply the multi-di-
mensional method to the economic content words instead of the functional features 
of typical MDA register analyses; this is similar to other recent developments in 
applying a multi-dimensional approach on content words (cf. e.g. Zuppardi/Berber 
Sardinha 2020). We also combine our textual data with metadata from the English 
Short-Title Catalogue (ESTC), which has been harmonized and augmented by the 
Helsinki Computational History Group (Lahti et al. 2019).
This novel approach does not come without problems, however. As an example, the 
digitized texts in ECCO have been automatically converted from scanned images into 
plain text, and this optical character recognition is riddled with errors impacting both 
precision and recall (Hill/Hengchen 2019). This pilot study aims to be a step towards 
a systematic quantitative assessment of economic language use based on these large-
scale collections, taking a critical view on the prevailing challenges in such analysis. 
With the combined expertise of the linguists and historians in our group, we will shed 
new light on the development of economic discourse by producing a description that is 
better couched in historical linguistics than earlier claims of an “economic enlighten-
ment”, and pave the way for further research utilizing historical language data in the 
digital humanities.
355
2 BACKGROUND
The increased written use of vernacular languages (as opposed to the cosmopolitan 
Latin) was an uneven process, in the sense that some local languages became print lan-
guages earlier than others and some registers or genres were more prone to vernacu-
larization. In the medieval period, devotional literature comprised the majority of ver-
nacular manuscripts in Europe, and the vernacular sermon, for instance, required the 
learned to communicate with lay people (Crossgrove 2000, Muessig 2010: 267). The 
fields of science, technology, and medicine also had a significant vernacular readership 
already in the Middle Ages (Crossgrove 2000). Religion and medicine, in particular, 
were areas of life where the educated mixed with the uneducated, which may have in 
part prompted vernacularization in these domains. Similarly in the eighteenth century, 
the insight of needing to reach a larger proportion of the population was particularly 
present in discussions about economic improvement. As states were increasingly seen 
as being in economic competition with one another (Hont 2005), the application of 
economic reform demanded bridging academic knowledge with the everyday prac-
tices of civic organizations and concrete economic policy. Economic matters became 
at the same time more specialized as they became the object of increased academic 
inquiry, and accessible to a broader readership as key actors increasingly saw a need 
to not only produce new knowledge, but also disseminate it to practitioners outside 
the sphere of academia. 
The specialization of this knowledge is best visible in several chairs containing 
the word economy in their titles being established at European universities during the 
course of the eighteenth century (Magnusson 1992: 249–257) and an accentuated the-
oretical debate with treatises reasons and nature of prosperity in different European 
countries. Some volumes, like Plumard de Dangeul’s Remarques sur les avantages 
et desavantages de la France et de la Grand-Bretagne (1754), very concretely high-
lighted the economic rivalry between Great Britain and France, making them of great 
interest to the reading public, but most probably served to disseminate economic vo-
cabulary. As debates about the economy crossed borders, so did the terminology, which 
is perhaps reflected in the large portion of French borrowings in the new eighteenth-
century economic vocabulary in English.
In more practical terms, so-called economic or improvement societies that prolifer-
ated all over Europe in the eighteenth century (Stapelbroek/Marjanen 2012) ventured 
into new avenues to put new knowledge, instruments and models into practice. The 
Honourable Society of Improvers founded in Scotland in 1723, which is generally re-
garded as the first of such societies, not only gathered information for discussion among 
its members, but actively published accounts of how to improve crops and manage 
farms in the newspapers (Bonnyman 2012: 37–38). Often written communication was 
not enough. In disseminating knowledge of new technology, The Dublin Society for 
the Improvement of Husbandry, Agriculture and other Useful Arts (founded in 1731), 
not only distributed models and drawings of ploughs to be used with different crops and 
types of soil, but also started promoting ploughing matches to promote efficient new 
agricultural methods (Livesey 2012: 69).
356
All this required both new vocabulary and new publication channels through which 
this new terminology could reach newly relevant strata in society. Overall, the role of 
the dictionary as a platform for lexical innovation becomes evident from Chambers 
onwards.. For instance, encyclopaedia entries on economic matters multiplied from 
Diderot and d’Alembert’s Encyclopédie from the mid-century to Panckoucke’s Ency-
clopédie Méthodique published toward the end of the century (Shovlin 2006: 2–5). 
Many seventeenth century or early eighteenth century neologisms, such as political 
economy, commercial, consumption, and industry, spread in their use during the course 
of the eighteenth century (McIntosh 2020: 163–165). Many of these terms expanded 
the horizon of economic thinking from the practical to the more theoretical and further 
from local circumstances to a national (or even international) perspective. The remark-
able proliferation of eighteenth-century dictionaries listed in the English Short Title 
Catalogue is striking.
Economic vocabulary also penetrated more popular styles and registers. This can 
perhaps be seen in the orthography of some words, economy being the most obvious 
case. The form ‘economy’ overtook the forms ‘oeconomy’ and ‘œconomy’ by the end 
of the eighteenth century, marking a shift from spelling the word in the “classical” to 
a modern “English” way. Such orthographic changes stemmed from the gradual crea-
tion of local standards, printing technology and beliefs of what would serve the public 
best. In the 1759 book The abecedarian John Yeomans argued for dropping the ligature 
spelling for the word ‘economy’ and other foreign words as “we refuse to gesticulate 
the modes of spelling by any nations upon earth” (Yeomans 1759: 48). While some 
other suggestions, such as writing ‘se’ instead of ‘sea’, ‘te’ instead of ‘tea’ or ‘æful’ 
instead of ‘awful’, did not catch on, the overall aim of remedying “the present condition 
of our vulgar tongue” and making “learning easy” (Yeomans 1759: 15) was a driving 
force in reforming spelling.
The vernacularization of economic language can be seen as taking place in the inter-
section of new fora, new ideas about how to communicate economic knowledge, and the 
spread of both new and established economic vocabulary. Our focus lies on the last one 
of these, but our analysis is grounded in the idea that linguistic innovation was couched 
in the institutions, publication practices and the interplay between different registers that 
helped produce new lexical items as well as make them part of everyday language.
3 MATERIAL AND METHODS
3.1 Material
The dataset we employ is based on texts from the Eighteenth Century Collections On-
line (ECCO), which contains full texts of some 200,000 documents as computer-read-
able texts produced by optical character recognition (OCR). These documents have 
been linked to corresponding metadata from the English Short Title Catalogue (ESTC) 
by the Helsinki Computational History Group (COMHIS); this metadata has then been 
further refined and enriched (Tolonen et al. 2022). As our focus is on the spread and 
change of economic language, we have chosen to further focus on a specific subset 
of documents: records labelled in the ESTC as belonging to the Goldsmiths’-Kress 
357
Library of Economic Literature (GKL), a microfilm collection of economic literature 
in a broad understanding of “economic” (Whitten 1978).
When building our dataset based on this collection, we only included the earliest 
edition of any given work, since any reprints of older works will likely include linguis-
tic features characteristic of their original period of publication, and thus may confuse 
diachronic analyses. Taking also into account the fact that not every ESTC record is 
represented in ECCO, we ended up with a dataset of roughly 5,000 ESTC records 
linked to ECCO documents, down from the full 11,000 ESTC record GKL set. Because 
an ESTC record can map onto multiple documents in ECCO (e.g. in the case of multi-
volume works), this set of roughly 5,000 ESTC records corresponds to a set of about 
5,200 ECCO documents. Figure 1 shows the number of documents over time divided 
by type. We can see that the number of pamphlets included in our data is particularly 
high in the early decades of the century. This is related to the 1707 Union of England 
and Scotland, which also led to an increase in pamphleteering on economic topics. 
Although the document count of pamphlets in GKL in the 1710s is high, this did not 
have an impact on our analysis due to both the low word-count of the pamphlets and, 
more importantly, the fact that the Union debate did not include the incoming economic 
vocabulary that turned out to be central for this study. 
Figure 1. Number of ECCO documents in the GKL over time, divided by document type (pam-
phlet = <32 pages, in-between = 32–128 pages, book = >128 pages)2
2 About these different document types, see Mikko Tolonen, Eetu Mäkelä, and Leo Lahti, “Supplementary 
Information to The Anatomy of Eighteenth Century Collections Online (ECCO) article,” Supplement to 
‘Anatomy of ECCO,’ June 2022. https://doi.org/10.5281/zenodo.6683914
358
As the starting point of our analysis of economic lexis, we take the ‘trade and fi-
nance’ section of the Historical Thesaurus of the Oxford English Dictionary (HT). 
Edited by a team led by Professor Christian Kay at the University of Glasgow, the 
HT describes the semantic development of English based on the second edition of the 
Oxford English Dictionary (OED) and A Thesaurus of Old English. It rearranges the 
OED by meaning into hierarchical semantic classes under the top-level categories of 
‘the mind’, ‘society’ and ‘the external world’. Out of the 800,000 words and meanings 
included in the thesaurus, 14,989 distinct lexical items can be found under the category 
of ‘society > trade and finance’.
Thanks to an agreement with Oxford University Press, we have access to local 
XML versions of the OED and HT. This enables us to easily retrieve not only a list 
of all words belonging to a HT category but also the lexicographical metadata associ-
ated with each word, such as etymology type (e.g. borrowing or derivative), etymon 
language (e.g. Latin or English), part of speech, year of first attestation, and the specific 
HT class of the word (e.g. ‘society > trade and finance > illegal or immoral trading > 
trade in (goods) illegally or immorally > smuggle’).
Using the OED as a source for first attestation dates in particular is not unproblem-
atic (cf. Durkin 2002). Work on the third edition is ongoing, and there are still entries in 
the dictionary that remain essentially unchanged since the first edition, prepared in the 
late nineteenth and early twentieth centuries. It is very probable that some of the entries 
have been antedated since March 2019, when we acquired the data, or will be antedated 
in the near future. Nevertheless, our usage of the OED is not dependent on the dating 
being absolutely correct. We are interested in economic lexis, whether old or new, that 
was extant in the eighteenth century and that increased in frequency during that century 
in ECCO. To make sure that we are not missing words with economic senses that were 
extant in the eighteenth century but that the OED dates somewhat later, we take all 
words in the HT ‘trade and finance’ category with an OED first attestation date starting 
from Old English and ending in 1850. The cutoff of 1850 rather than 1800 accounts for 
possible antedatings to our OED data. In our investigation of specific lexical items as-
sociated with the dimensions of incoming economic vocabulary identified in our analy-
sis (4.1 below), we retrieved the first attestation dates from OED Online in spring 2022. 
While the same caveats apply, the datings are used as indicative of the relative age of 
the words, particularly across the different dimensions, and our main findings regarding 
the dimensions do not rely on exact dates.3
The OCR of the ECCO data used is not perfect and causes some self-evident prob-
lems in matching strings of characters. In general, the distribution of errors is relatively 
even so that the results of most of our quantitative analyses are not affected by poor 
OCR. There are, however, some exceptions to this: the use of the long ‘s’ in eighteenth-
century documents, as well as the use of ligatures, both of which often include errors in 
the machine-readable version (Hill/Hengchen 2019). We manually inspected the OCR 
3 In analyses which focus more heavily on the dates of introduction of new lexical items, the OED data could be 
supplemented by e.g. lexical lookups of the full ECCO dataset and other datasets such as the British Newspaper 
Archive.
359
quality of the spelling variants ‘oeconomy’, ‘œconomy’ and ‘economy’ in a selection 
of texts, and could confirm that about a third of the machine-readable strings ‘econ-
omy’ were, in fact, false renderings of the form ‘œconomy’. Regardless, our manual 
inspection shows that the spelling variant ‘economy’ did become more common in the 
last decades of the eighteenth century, but any detailed quantitative assessment of spell-
ing variation would not be reliable without manual inspection due to ligatures being 
exceptionally prone to errors. Based on this inspection, we also note that results with 
regard to words with the letter ‘s’ potentially include a higher rate of errors.
3.2 Methods
In order to study the changes in economic vocabulary, we make use of a two-phase 
methodology. In the first phase, we identify incoming economic vocabulary; in the sec-
ond phase, we identify underlying discourses driving the rise of some of these words.
3.2.1 Incoming economic vocabulary
We start our analysis of economic vocabulary with the set of lexical items classified 
under ‘trade and finance’ in the HT. As our focus is on the eighteenth century, we 
further limit this set to those items which are first attested in the ‘trade and finance’ 
category before the year 1850. We set this limit to avoid including items which were 
used with other meanings in the eighteenth century but which only later developed an 
economic meaning, but offset the limit by 50 years because it is possible there may be 
antedatings in our dataset of the first attestations recorded in the OED, as discussed 
in 3.1 above. We also exclude items whose entries contain parenthetical remarks or 
alternative constructions, which would not appear in that form in running text. After 
the set of items was determined, the frequency of each of the 7,079 items remaining 
in this set was calculated by decade by dividing the total number of occurrences of the 
item in each decade of our data by the total number of characters4 in the OCR versions 
of the documents.
To identify items which rise in frequency relatively consistently throughout the 
century, we calculated the proportional change rate (PCR; Nevalainen et al. 2020) of 
each of the items across the decades. Originally devised to help identify time periods 
with the highest rate of change, the PCR describes how much of the total change in the 
frequency of a lexical item happens between each pair of successive time periods. In 
order to focus on incoming items with relatively constant growth, we excluded all items 
which had a PCR higher than 10% during any period of negative growth, that is to say, 
items which had at least one period of decreasing frequency which constituted over 
10% of their total frequency changes. To avoid spurious results and statistical noise, 
we further excluded all items which had fewer than 100 total occurrences throughout 
our dataset, as well as items which did not appear in at least two decades. In the end, 
the set of lexical items of potential interest included 73 items. Figure 2 shows that the 
4 Even though using the word count of the documents as the normalization base would be preferable, we use 
the character count instead to mitigate the effects of the questionable OCR quality, which makes it difficult to 
determine the word count of a document.
360
increase in frequency happened dominantly in the latter half of the century, with the 
most common words being central in economic vocabulary.
Figure 2. Normalized frequency of top 8 words among the 73 consistently increasing words 
identified by the PCR analysis
3.2.2 Underlying discourses
In order to identify some of the discourses which may underlie these incoming lexical 
items, we employ factor analysis, inspired by the multi-dimensional method of register 
analysis (e.g. Biber 1988). Factor analysis identifies “a set of latent constructs underly-
ing the battery of measured variables” (Fabrigar 1999), in our case, a set of discourses 
underlying the use of certain incoming lexical items. In practical terms, we can use 
factor analysis to identify groups of lexical items which vary in frequency together, 
that is, whose frequencies in documents are correlated, so which tend to have a higher 
frequency in a document or a lower frequency in a document at the same time. The 
assumption behind the MDA methodology is that there is a functional reason why 
certain features tend to co-occur in the same documents. In our case, we interpret the 
co-occurrence of certain items of economic vocabulary to be because of their shared 
contexts of use, and therefore because of certain underlying discourses (cf. e.g. Zup-
pardi/Berber Sardinha 2020).
The factor analysis as performed here draws heavily on the principles and solu-
tions used by Biber (1988). Factor analysis can be performed using various factoring 
methods; we use the commonly-used Principal Axis Factoring. In order to make the 
extracted factors more readily interpretable, the resulting factor solution needs to be 
rotated, i.e. the axes of the factor space are changed so that separate groups of co-
occurring items load more strongly on separate factors. We did this using the oblimin 
rotation method, which allows for a degree of correlation between the extracted factors. 
361
Allowing potential correlations is desirable because it is not reasonable to assume that 
the underlying discourses are completely independent from each other. The factor anal-
ysis itself was run twice. In the first round, five factors were extracted, which was the 
highest number of factors that could be extracted without issue, to catch as much of the 
overall variation as possible. After this round, all items which did not have a loading 
equal to or higher than |0.2| on any of the factors were removed, as such items were 
not very important for the description of any of the factors. Only the 28 items left over 
were included in the second round of the analysis. In this round, the highest number 
of factors that could be extracted without issue was four. All factor solutions of two to 
four factors were investigated. In the end, the four-factor solution was chosen because 
it explains the highest proportion of the overall variation and all four factors appeared 
readily interpretable.
Next, dimension scores were calculated for each ESTC record in our dataset. First, 
each lexical item was assigned to the factor(s) it had loadings over |.30| on. Then, the 
frequencies of each of the items were standardized to their mean and standard devia-
tion. This was done to prevent high-frequency items from drowning out the effect of 
low-frequency features. After this, the dimension score of each ESTC record on each of 
the dimensions was calculated by summing the standardized frequencies of the features 
associated with that dimension within that document.
In our analysis of the functions and underlying discourses of the dimensions, we 
make use of standard corpus-linguistic methods such as concordance lines and colloca-
tion analysis; the results of this analysis are reported in subsection 4.2 below.
4 ANALYSIS
Following the procedure described above, we extracted the four dimensions displayed 
in Table 1. The items on each dimension tend to be present in the same texts and absent 
in the same texts, but the different dimensions are free to vary independently from each 
other. In practice, there is a small degree of correlation between the different factors, 
varying between 0.08 and 0.37, depending on the pair of factors. In this section, we 
will first look into the individual items associated with these dimensions in subsection 
4.1. Then, in subsection 4.2, we will analyze the dimensions to interpret, define, and 
label them.
Table 1. The four extracted dimensions and the lexical items associated with them
Dimension Lexical items
D1 Income, expenditure, incidental
D2 Circulating medium, funded, unfunded
D3 Income, finance, financier, financial, funding
D4 Commercial, liberal, improvement, extent, produce
One thing of note is that, unlike in many multi-dimensional studies, the above di-
mensions only include lexical items with high positive loadings, and no items with high 
362
negative loadings. The items with positive and negative loadings on a dimension tend 
to be in a complementary distribution: when the frequency of the positive items is high, 
the frequency of the negative items is low, and vice versa. In other words, none of these 
groups of lexical items is associated with a contrasting complementary group of items. 
The reason for this is largely technical: overall, the items included on these dimensions 
are relatively rare in the full dataset, which means that the most common frequency for 
these items in the documents is zero. Due to this, a decrease in the frequency of any 
set of items can not be easily correlated with an increase in the frequency of any other 
complementary set of items. However, the co-occurring groups of words in these four 
dimensions can already tell us a great deal about the kinds of discourses in which these 
rising lexical items were used.
In subsection 4.1, we will first explore the lexicographical metadata related to the 
HT ‘trade and finance’ words more generally. Then, we will focus on the incoming 
items included in the four extracted dimensions and explore the lexicographical meta-
data of these items in more detail. In subsection 4.2, we will then analyze the four 
dimensions, interpreting the discourses underlying the dimensions extracted in the sec-
tion above using standard corpus-linguistic methods.
4.1 Lexicographical metadata
We combined the HT lemmas from the ‘society > trade and finance’ category with 
OED metadata. While the ID correspondences between the HD and OED are not com-
plete, this yielded us a set of 4,084 trade and finance lemmas which could be combined 
with their metadata. We were thus able to examine the source languages of new trade 
and finance lemmas in conjunction with the dates of their first attestation. It should be 
noted that once a word has entered the English language from another language (e.g. 
commercial, dated ante-1687, source language Latin), its subsequent derivatives (e.g. 
commercialism, dated 1849) have English as their source language. Thus English is the 
source language not only of native neologisms but also of a substantial group of lemmas 
whose roots may ultimately be in other languages. Of the 4,084 lemmas in this dataset, 
with first attestations between 601–4 and 1994, 2,498 have English as their source lan-
guage. The second-largest source language is French (524 lemmas in total), followed 
by Latin and German (290 and 59 lemmas, respectively). The number of new words 
per century rises to its highest point in the 1600s (813 lemmas) and actually drops in 
the 1700s (443 lemmas), but rises again in the 1800s (762 lemmas). French was the 
most prolific foreign source of new words in the fourteenth and fifteenth centuries (102 
and 125 new lemmas, respectively), which is to be expected, considering the position 
of French in medieval Europe as “the language of the princely courts and the courts of 
law, of high culture (secular and religious), and of bourgeois aspiration and trade” (Put-
ter and Busby 2010: 3). Latin reached its peak as the most common non-English source 
language in the seventeenth century (122 new lemmas), at a time when Latin was still 
a popular publishing language but vernacular publishing had already surpassed it in 
Britain and was starting to grow substantially (Marjanen et al. Under review).
363
Figure 3. The source languages and first attestation dates of HT ‘trade and finance’ lemmas in 
languages with over 50 occurrences based on OED metadata. English lemmas include words that 
were derived within English from lemmas of foreign origin
We further analyzed the OED and HT metadata on the lexical items associated with 
each dimension. In Dimension 1, all words belong to the HT class ‘society > trade and 
finance > management of money’. While income was first attested in 1601 according 
to the OED, both expenditure and incidental (as in incidental charge/expense) are new 
to the latter half of the eighteenth century, first attested in 1769 and 1791, respectively. 
Both words are somewhat learned derivatives: expenditure was derived from Latin 
expenditus + -ure, while incidental comes from incident + -al, perhaps modelled on 
French incidentel. We may thus expect these items to co-occur in rather specialized 
registers discussing the management of money, chiefly from the 1790s onwards.
The words in Dimension 2 are found under ‘money’, ‘management of money’, and 
‘financial dealings’ within the trade and finance category of the HT. While funded and 
unfunded were first attested in the 1760s and 1770s, respectively, the OED gives cir-
culating medium a first attestation date as late as 1803, although the compound was 
clearly in use earlier than that based on our dataset. The noun fund was borrowed from 
Latin (fundus); the verb fund as well as the adjectives funded and unfunded were de-
rived within English. Circulating likewise stems from the Latin circulāt, and medium 
is a direct borrowing from Latin.
Dimension 3 shares the word income with Dimension 1; the noun funding, first at-
tested in 1735, shares the same root as funded and unfunded in Dimension 2. The rest of 
the words in Dimension 3 are interrelated: finance, financier and financial can be traced 
back to Old French finance, although financier is a later borrowing of the French word. 
Finance is given various categories and meanings in the HT, with first attestations 
364
ranging from 1439 to 1866. Financier is first attested in the 1600s and financial in the 
late 1700s. The Dimension 3 words mostly appear in the ‘financial dealings’, ‘man-
agement of money’, and ‘fees and taxes’ subcategories of trade and finance, although 
finance has entries under ‘money’ and ‘payment’ as well.
Apart from commercial, the words in Dimension 4 have a considerably higher num-
ber of meanings in other HT categories than they do in trade and finance. Commer-
cial, improvement, and extent were first attested in the late 1600s, 1400s, and 1300s, 
respectively, and their first trade and finance senses date back to the same periods. 
Liberal, however, was first attested in the late 1300s but its recorded trade and finance 
usage only dates back to 1816, even if the OED does give an example of the economic 
sense of the word developing in the 1770s; produce was first attested in the 1400s and 
had its trade and finance first attestation in 1585. As for the origins of the Dimension 
4 words, produce was borrowed from Latin and improvement partly borrowed from 
French, partly derived within English, while commercial, liberal, and extent have a mix 
of French and Latin roots that, in the case of liberal and extent, date back to the middle 
ages. Overall, these words are older than the other dimension words, and have more 
meanings outside the economic.
It seems that the trade and finance vocabulary that increased in frequency during 
the eighteenth century is comparatively young and of French or Latin origin. The first 
attestations listed in the OED, whether for the word overall or for a particular usage, 
are of course not precise and may change when new sources are discovered, but they 
are a reliable indication of the dates by which the word or usage appears at the latest. 
We also looked at trade and finance lemmas whose frequency decreases or stays the 
same over this period and discovered that many of them, in contrast, are inherited from 
Germanic and date back to Old English – for example star, fly, and sit. However, they 
also tend to have several meanings that are not directly related to economic topics and 
somewhat rare or obscure trade and finance usage. A closer look at handpicked stable 
or decreasing trade and finance related words such as traded, imperial, and pay – the 
first two of multiple origins and pay a borrowing from French – reveals that even the 
decreasing ones tend to be of mixed origin, but they are typically older borrowings than 
the increasing dimension words above – ounce, for example, is partly a borrowing from 
Latin and dates back to Old English.
4.2 Analysis of the dimensions
A central means of understanding the dimensions is to analyze how the lexical items as-
sociated with the dimension are actually used in context. In practice, this is done using 
concordance lines and close reading of the texts. To start with, we searched for each of 
the items in three highly scoring texts per dimension from the 1790s, the period of most 
change in our data. Dimensions 1 and 3 share a lexical item (income) and also share one 
of the texts in this selection. Of the lexical items of Dimension 2, especially expendi-
ture but also the shared item, income, were repeated frequently as column headings or 
labels in tables appended to the texts, which may explain their frequency and charac-
terize this dimension. All three texts inspected for Dimension 1 dealt with income tax 
365
as a means to solving state expenditure and national debt (there was only one instance 
of the third lexical item, incidental, in the three texts). The cause for said debt is also 
the focus of Dimension 2, in which the bills and debts that are funded or unfunded are 
identified as those of the navy in particular, caused by the American revolutionary war 
(circulating medium appears more rarely but is also related to the need to manage state 
finances). Notably Dimension 2 consists almost entirely of anonymous pamphlets; the 
texts surveyed here are by known authors. 
The topics of Dimension 3 are much the same as 1 and 2 but with a focus on finan-
cial measures; funding appears repeatedly in the context of “the funding system”, also 
referred to as “the English System of Finance”, so this dimension seems to capture an 
awareness of the systemic change underway. Dimension 4 stands out as the one that 
deals with commerce, agricultural production, manufacture and industry, and the de-
velopment of and relationships between particular (often colonial) areas.
Since it seems based on the close reading of a selection of the texts that the di-
mensions are meaningful and interpretable, we performed a larger-scale analysis of 
the dimensions using standard corpus-linguistic methods, most importantly collocation 
analysis and concordance lines. To do this, we created a subcorpus of all of the texts 
which are in the top 5 per cent on any of the dimensions. We then made use of AntConc 
(Anthony 2022) to analyze the collocations and concordance lines of all of the words 
included in the dimensions. Specifically, we searched for collocations within a window 
of five in both directions, and which appear at least in five separate texts and five times 
in total in the subcorpus.
This analysis is in line with the findings of the close reading. Dimension 1 appears 
to deal particularly with public income and expenditure. In the top document subcorpus, 
the words income and expenditure collocate with words such as national, taxable, an-
nual, and public. Incidental is commonly used to refer to additional expenses or situa-
tions which may cause them, such as “other circumstances incidental to war”; indeed, 
war appears as a collocate of incidental in the analysis. Dimension 2 has to do with 
public debt. As in the close reading above, funded and unfunded both overwhelmingly 
relate to funded and unfunded public debt. Dimension 3 is related to the financial system 
as a whole, for instance, politics, committees, ministers, reports, and resources related to 
the financial system. Dimension 4 refers to various aspects of private commercial, agri-
cultural, and manufacturing enterprise. Commercial is often used to refer to commercial 
actors as opposed to e.g. the landed interests, or e.g. commercial treaties; improvement is 
used most commonly in reference to the improvement of e.g. an extent of land.
To corroborate this analysis of the functions of the dimensions, we also looked 
at the authors and titles of the ten highest-scoring documents from each dimension. 
The results of this analysis also closely mirror the above analyses. The documents 
on Dimension 1 overwhelmingly deal with public finances and income, particularly 
the income tax. This dimension includes texts such as “Observations on the produce 
of the income tax, and on its proportion to the whole income of Great Britain” by 
Henry Beeke and “A review of Dr. Price’s writings, on the subject of the finances of 
this kingdom” by William Morgan. The texts on Dimension 2 either have to do with 
366
public debt and credit, or the state of the nation’s finances more generally. These 
include texts such as “An inquiry into the state of the finances of Great Britain” by 
Nicholas Vansittart and “Observations on the national debt, and an enquiry into its 
real connection with the general prosperity” by an unknown author. The titles on Di-
mension 3 are less focused, but many of them are still in various ways related to the 
system of finance and national funding, such as “A letter on the present measures of 
finance” by James Maitland Lauderdale or “A plan for raising the supplies during the 
war” by an unknown author. Dimension 4 clearly deals with trade, commerce, manu-
facture, and the improvement of land. These works include titles such as “A represen-
tation concerning the knowledge of commerce as a national concern; pointing out the 
proper means of promoting such knowledge in this kingdom” by J. Massie, and “A 
dissertation on the chief obstacles to the improvement of land, and introducing better 
methods of agriculture throughout Scotland” by an unknown author. In general, the 
evidence suggests the significance of Scottish influence in shaping the development 
of economic discourse (see Hont 2005). The majority of the authors producing the ten 
highest-scoring documents for each dimension are Scottish. There is a clear practical 
element of improvement present also in these texts that can be seen as the dominat-
ing aspect of the impact of the Scottish Enlightenment on the economic discourse in 
eighteenth-century Britain (cf. Sher 1985).
Figure 4. The average dimension scores per decade in the ECCO GKL documents for all four 
dimensions
367
However, since we are focusing on incoming lexical items, we may expect there to 
also be diachronic differences in their contexts of use. Since all of the items included 
on the dimensions are much more common in the later decades of the century, their fre-
quent appearance may overshadow their use in earlier periods in an analysis which does 
not take the diachronic dimension into account. Figure 4 shows the average dimension 
scores of all GKL documents included in ECCO for each decade of the century. This 
figure shows the effects of focusing on incoming items very well, as the documents in 
the latter decades of the century have considerably higher average dimension scores 
than the documents in the earlier periods due to the overall higher frequency of the 
items towards the end of the century. We can also see in the figure how Dimension 4, 
which includes the earliest attested words of any of the dimensions, increases through-
out the century; on the other hand, Dimension 1 increases slowly first but then rapidly 
at the end of the century, and Dimensions 2 and 3 are all but nonexistent before the end 
of the century (cf. Table 2).
In order to extend the analysis to the diachronic changes in the contexts of use of 
the lexical items, we created four additional subcorpora, one for every quarter-century, 
which included the texts which are in the top 5 per cent on any of the dimensions dur-
ing that quarter-century. Table 2 summarizes our diachronic findings for each of the 
dimensions. Based on the analysis of these subcorpora, in the beginning of the century, 
generally speaking, the items tend to be used with more concrete meanings and in more 
concrete contexts. Towards the end of the century, their use diversifies, and the use of 
the items in more abstract and general senses increases alongside the more concrete 
ones. For instance, income on Dimensions 1 and 3 first often refers to the income of cer-
tain groups of people or e.g. estates or specific funds. Towards the end of the century, 
more references start appearing to higher-level and larger-scale issues such as national 
income and expenditure, including e.g. income tax, which was introduced in the UK 
around this time. Similarly, in the beginning of the century, produce on Dimension 4 
often refers to what is produced in concrete terms by estates, lands, or the country. Its 
use also diversifies over the century, and it too is used more and more to refer to more 
abstract concepts such as produce of taxes, of the customs, of the state, or of labour. Of 
course, these patterns are only tendencies: examples of both concrete and abstract uses 
can be found from all time periods a word appears in. Some manifestations of these 
observed patterns are illustrated by Examples 1 and 2 below. Example 1 shows the 
word income referring very directly to the income of a certain set of gentry, whereas 
in Example 2 the words income and expenditure refer to the financial situation of the 
country as a whole.
(1) There is hardly one of this new Set of Gentry, from Two Thousand Pound 
Fortune and upwards that do not spend near half their Income in foreign 
Wines, Linens, Silks, Laces, Tea, Coffee, and an infinite number of other 
Curiosities.
Anonymous (1720): Considerations on the present state of the nation
368
(2) [H]e has consequently increased all the other property in the kingdom, if 
not precisely in the same proportion, certainly to a considerable amount: he 
has not only made the income of the country equal to its expenditure, but 
has also procured a surplus of a million per annum, to be employed in the 
reduction of the national debt.
George Pretyman (1786): A short answer to Earl Stanhope’s observations on 
Mr. Pitt’s plan for the reduction of the national debt
The word commercial on Dimension 4 also starts out as referring mostly to more 
concrete groups and actors, such as “trading men in the commercial world”. As the 
century progresses, the number of references to the country and its people as a commer-
cial actor increases, including collocates like “a commercial nation”, “a commercial 
people”, or “commercial empire”, reflecting a view of the nation as a whole basing its 
identity on commerce. Other higher-level uses of commercial increase through the lat-
ter half of the century, such as “commercial intercourse”, “commercial treaties”, “com-
mercial concerns”, or “commercial system”. These changes start taking place earlier 
than most of the others, as suggested by the earlier increase in the graph in Figure 4, and 
they point to the emergence of a more specialized and abstract discourse with regard 
to political economy. Economic theory addressed commercial activity as increasingly 
international and was less engaged in private entrepreneurship.
Table 2. A summary of the main diachronic tendencies in the contexts of use of the words as-
sociated with the four dimensions
Dimension Beginning of the century End of the century
D1 public income and 
expenditure: income, 
expenditure, incidental
Mostly private income of 
individuals or companies
Both private and public 
income and expenditure, and 
the relationship between the 
two, e.g. taxation of income
D2 public debt: circulating 
medium, funded, unfunded
Very rare Mostly funded and unfunded 
debt, also e.g. funded 
property or funded taxes
D3 financial system: 
income, finance, financier, 
financial, funding
Very rare Various aspects of the 
financial system and its 
public uses, including the 
funding of debts, bills, or the 
military
D4 private enterprise: 
commercial, liberal, 
improvement, extent, 
produce
Commerce, agriculture, 
manufacturing, etc. as an 
activity of the people
Commerce also as an 
international activity and a 
matter of national interest
369
In general, these changes appear to reflect the overall development of the system of 
economy over the century. For example, with the exception of income, all of the lexi-
cal items associated with Dimension 3 – finance, financier, financial, funding – were 
extremely rare in the first half of the century, with an increasing number of instances 
from the third quarter of the century onwards. 
Furthermore, some of the diachronic differences represent topics which have been 
of particular interest or importance in specific periods. The most prominent of these 
is the inclusion of military matters, most importantly “navy” and “war”, in economic 
discourse particularly in the final quarter of the century. This vocabulary mirrors the 
economic competition that concretely led to war efforts in the years after the American 
and French revolutions.
5 DISCUSSION
This paper has focused on incoming economic vocabulary extracted from a set of words 
based on previous lexicographical research. These choices have naturally excluded a 
number of potentially interesting words whose frequency may not have consistently 
increased in the dataset as a whole or that may not have been identified as relevant to 
economic discourse in the sense of the ‘trade and finance’ section of the HT. There is 
therefore much to be done in future research on eighteenth-century English economic 
lexis. One avenue would be to identify relevant words in a more data-driven manner 
based on the corpus alone. Of course, the GKL itself does not represent all of eight-
eenth-century economic discourse, so we could also augment the corpus by identifying 
similar texts in the full ECCO (cf. Tiihonen et al. 2022). As mentioned in Section 3.1 
above, OCR errors are an issue in the data that is yet to be fully resolved, although us-
ing the number of characters as the normalization base has mitigated its effects some-
what. Recall could be improved by using fuzzy searches, whereas improving precision 
could require manual consultation of the document images, as discussed with regard to 
the spelling variants of economy in Section 2.
In the previous section we identified lexical dimensions that grouped together in-
coming economic vocabulary co-occurring in eighteenth-century texts in the GKL. It 
would be of interest to relate these dimensions to different registers in the corpus: for 
instance, parliamentary debates could be an important register in the development of 
discourses of public economy (dimensions 1 and 2), and legal texts may have helped to 
spread specialist vocabulary to wider public consciousness. In future research, we aim 
to generate register information for GKL texts using both Biber’s linguistic features 
and machine learning methods that would identify registers through extrapolation from 
existing corpora with register metadata. This would also enable us to conduct a more 
thorough investigation into how Biber and Finegan’s (1997) finding of the divergence 
of popular and specialized registers from the eighteenth century onwards relates to the 
domain of economy and economic vocabulary.
In general, our analyses showcase how economic vocabulary grew in frequency and 
went through a lexical diversification. We hypothesized that this development would 
be related to an expansion of registers and the diffusion of economic discourse to more 
370
popular forums as well as increased intellectual inquiry. However, our analyses only 
support the latter part of the hypothesis, indicating that economic terms started to appear 
in increasingly abstract uses and that a more specialized economic discourse emerged 
during the eighteenth century. The trends of a democratization and a specialization of 
economic discourses are not mutually exclusive, but perhaps our analysis, with its fo-
cus on incoming rather than established vocabulary, is more prone to capture the latter. 
The specialization is clearly related to the so-called economic bestsellers in eighteenth-
century Europe (Carpenter 1975) and the consequent emergence of political economy 
as a field for intellectual theorizing. The obvious example of this is the lexicalization of 
the term political economy itself.
6 CONCLUSION
We investigated incoming economic lexis and the discourses underlying their rise in 
the eighteenth century in a collection of economic literature. We analysed the lexical 
metadata of economic vocabulary and found that the incoming economic vocabulary is 
largely Latin or French in origin, whereas the stable and outgoing economic vocabulary 
tends to be either of native English Germanic origin or older loans from e.g. French or 
Dutch, with dominant non-economic meanings. In order to identify incoming lexical 
items, we made use of PCR (Proportional Change Rate), a novel method for identifying 
periods of linguistic change and items of interest.
Using multi-dimensional analysis methods inspired by Biber (1988), we extracted 
four dimensions of incoming economic lexis. While similar methods have been applied 
to content words before (e.g. Zuppardi/Berber Sardinha 2020), we used the method for 
the purposes of identifying incoming economic discourses in historical textual data. 
We identified four lexical dimensions: public income and expenditure, public debt, 
financial system, and private enterprise. The development and rise of the lexis related 
to these dimensions can be linked to the overall development of both the economic 
system itself and the contemporary economic discourse in the eighteenth century. By 
analyzing actual usage in a large historical database of texts, our study has contributed 
new evidence for the hypothesis that lexical change often reflects broader sociocultural 
change (e.g. Allan 2015).
References
ALLAN, Kathryn (2015) “Education in the Historical Thesaurus of the Oxford English 
Dictionary: Exploring diachronic change in a semantic field.” In: Jocelyne Daems/
Eline Zenner/Kris Heylen/Dirk Speelman/Hubert Cuyckens (eds), Change of para-
digms – new paradoxes: Recontextualizing language and linguistics. Berlin: Walter 
de Gruyter, 81–95.
ANTHONY, Laurence (2022) AntConc (Version 4.0.5) [Computer Software]. Tokyo: 
Waseda University. https://www.laurenceanthony.net/software
BIBER, Douglas (1988) Variation across Speech and Writing. Cambridge: Cambridge 
University Press.
371
BIBER, Douglas/Edward FINEGAN (1997) “Diachronic relations among speech-
based and written registers in English.” In: Terttu Nevalainen/Leena Kahlas-Tarkka 
(eds), To explain the present: Studies in the changing English language in honour 
of Matti Rissanen. Helsinki: Société Néophilologique, 253–275.
BONNYMAN, Brian (2012) “Agrarian Patriotism and the Landed Interest: The Scot-
tish ‘Society for Improvers in the Knowledge of Agriculture’, 1723–1746.” In: Koen 
Stapelbroek/Jani Marjanen (eds), The Rise of Economic Societies in the Eighteenth 
Century: Patriotic Reform in Europe and North America. Basingstoke, Palgrave 
Macmillan, 26–51.
CARPENTER, Kenneth E. (1975) “The economic bestsellers before 1850: A catalogue 
of an exhibition prepared for the History of Economics Society meeting, May 21–
24, 1975, at Baker Library.” Bulletin of the Kress Library of Business and Econom-
ics, Harvard Business School 11, 1–33.
CROSSGROVE, William (2000) “The Vernacularization of Science, Medicine, and 
Technology in Late Medieval Europe: Broadening Our Perspectives.” Early Sci-
ence and Medicine, 5(1), 47–63.
CULPEPER, Jonathan/Minna NEVALA (2012) “Sociocultural processes and the his-
tory of English.” In: Terttu Nevalainen/Elizabeth Closs Traugott (eds), The Oxford 
handbook of the history of English. Oxford: Oxford University Press, 365–391.
DURKIN, Philip (2002) Changing documentation in the third edition of the Oxford 
English Dictionary: Sixteenth-century vocabulary as a test case. In: Teresa Fanego/
Bélen Méndez-Naya/Elena Seoane (eds), Sounds, words and change: Selected pa-
pers from 11 ICEHL, Santiago de Compostela, 7–11 September 2000. Amsterdam: 
John Benjamins, 65–81.
FABRIGAR, Leandre R./Duane T. WEGENER/Robert C. MacCALLUM/Erin J. 
STRAHAN (1999) Evaluating the use of exploratory factor analysis in psychologi-
cal research. Psychological Methods 4/3, 272–299.
HILL, Mark/Simon HENGCHEN (2019) “Quantifying the impact of dirty OCR on 
historical text analysis: Eighteenth Century Collections Online as a case study.” 
Digital Scholarship in the Humanities 34/4, 825–843.
HONT, Istvan (2005) Jealousy of trade. International Competition and the Nation-
State in Historical Perspective. Cambridge MA: Harvard University Press.
JONES, Peter (2016) Agricultural enlightenment: Knowledge, technology, and nature 
1750–1840. Oxford: Oxford University Press.
KLEINHENZ, Christopher/Keith BUSBY (eds), Medieval Multilingualism: The Fran-
cophone World and its Neighbours. Turnhout: Brepols,
LAHTI, Leo/Jani MARJANEN/Hege ROIVAINEN/Mikko TOLONEN (2019) “Bib-
liographic Data Science and the history of the book (c. 1500–1800).” Cataloging & 
Classification Quarterly 57/1, 5–23.
LIVESEY, James (2012) “A Kingdom of Cosmopolitan Improvers: The Dublin Soci-
ety, 1731–1798.” In: Koen Stapelbroek/Jani Marjanen (eds), The Rise of Economic 
Societies in the Eighteenth Century: Patriotic Reform in Europe and North Ameri-
ca. Basingstoke: Palgrave Macmillan, 52–72.
372
McINTOSH, Carey (2020). Semantics and cultural change in the British Enlighten-
ment: New words and old. Leiden: Brill.
MAGNUSSON, Lars (1992) “Economics and the Public Interest: The Emergence of 
Economics as an Academic Subject during the 18th Century.” Scandinavian Jour-
nal of Economics 94 (Supplement), 249–257.
MARJANEN, Jani/TUULI Tahko/Leo LAHTI/Mikko TOLONEN (forthcoming) 
“Book Printing in Latin and Vernacular Languages in Northern Europe, 1500–1800.”
MUESSIG, Carolyn (2010) “The Vernacularization of Late Medieval Sermons: Some 
French and Italian Examples.” In: Ch. Kleinhenz/K. Busby (eds), 267–284. 
NEVALAINEN, Terttu (1999) “Early Modern English lexis and semantics.” In Roger 
Lass (ed.), The Cambridge history of the English language, III: 1476–1776. Cam-
bridge: Cambridge University Press, 332–458.
NEVALAINEN, Terttu/Tanja SÄILY/Turo VARTIAINEN/Aatu LIIMATTA/Jefrey 
LIJFFIJT (2020) ”History of English as punctuated equilibria? A meta-analysis of 
the rate of linguistic change in Middle English.” Journal of Historical Sociolinguis-
tics 6/2, str. 1–40.
POPPLOW, Marcus (2010) “Economizing agricultural resources in the German eco-
nomic enlightenment.” In: Ursula Klein/E. C. Spary (eds), Materials and expertise 
in early modern Europe. Chicago, IL: University of Chicago Press, 261–287.
PUTTER, Ad/Keith BUSBY (2010) “Introduction.” In: Ch. Kleinhenz/K. Busby (eds), 
1–13.
ROBERTSON, John (2005) The Case for The Enlightenment. Scotland and Naples 
1680–1760. Oxford: Oxford University Press.
SHER, Richard (1985) Church and University in the Scottish Enlightenment: The Mod-
erate Literati of Edinburgh. Edinburgh: Edinburgh University Press.
SHOVLIN, John (2006) The political economy of virtue: Luxury, patriotism, and the 
origins of the French Revolution. Ithaca, NY: Cornell University Press.
TIIHONEN, Iiro/Yann RYAN/Lidia PIVOVAROVA/Aatu LIIMATTA/Tanja SÄILY/
Mikko TOLONEN (2022) “Distinguishing discourses: A data-driven analysis of 
works and publishing networks of the Scottish Enlightenment.” In: Karl Berglund/
Matti La Mela/Inge Zwart (eds), Proceedings of the 6th Digital Humanities in the 
Nordic and Baltic Countries Conference (DHNB 2022), 120–134. https://ceur-ws.
org/Vol-3232/paper09.pdf
TOLONEN, Mikko/Eetu MÄKELÄ/Leo LAHTI (2022) “The Anatomy of Eighteenth 
Century Collections Online (ECCO).” Eighteenth-Century Studies 56, 95–123. 
WHITTEN, David O. (1978) “Democracy returns to the library: The Goldsmiths’-Kress 
Library of Economic Literature.” Journal of Economic Literature 16/3, 1004–1006.
ZUPPARDI, Maria Carolina/Tony BERBER SARDINHA (2020) “A multi-dimension-
al view of collocations in academic writing.” In: Ute Römer/Viviana Cortes/Eric 
Friginal (eds), Advances in corpus-based research on academic writing: Effects of 
discipline, register, and writer expertise. Amsterdam: John Benjamins, 333–353.
373
Abstract
DIMENSIONS OF INCOMING ECONOMIC VOCABULARY IN EIGHTEENTH-
CENTURY BRITAIN
The eighteenth century is often connected with economic improvement. Considering 
the significant functional expansion of the English language during this period, driven 
by various socio-cultural changes, and the contemporary interest in the economy, we 
hypothesize that this linguistic expansion facilitated the spread of economic vocabulary 
to new contexts. Combining linguistic and historical expertise, we study vocabulary 
drawn from the ‘trade and finance’ section of the Historical Thesaurus of the Oxford 
English Dictionary in economic texts included in Eighteenth Century Collections On-
line. We identify incoming economic lexis based on its rate of change and apply multi-
dimensional analysis to extract four lexical dimensions of economic discourse, which 
we interpret as (1) public income and expenditure, (2) public debt, (3) financial system, 
and (4) private enterprise. The lexical items associated with the dimensions are mostly 
Latin or French in origin, and many of them are neologisms that are first attested in the 
later eighteenth century, suggesting their widespread introduction into the language 
around that time. We show that at the beginning of the century, the use of the items 
that were extant then tends to be more concrete and local, with more abstract and wide-
reaching contexts added towards the end of the century. This suggests a specialization 
of economic discourse that is related to the emergence of political economy as a field 
for intellectual theorizing.
Keywords: corpus linguistics, historical linguistics, multi-dimensional analysis, Eight-
eenth Century Collections Online, Eighteenth-century studies
Povzetek
ZNAČILNOSTI PREVZETEGA BESEDJA S PODROČJA GOSPODARSTVA V 
VELIKI BRITANIJI 18. STOLETJA 
18. stoletje pogosto povezujemo z gospodarskim napredkom. Ob upoštevanju takratne-
ga močnega funkcijskega razmaha angleščine, ki je bil posledica raznih družbeno-kul-
turnih sprememb, in sočasnega zanimanja za gospodarske zadeve domnevamo, da je 
ta jezikovni razmah pospešil razširitev besedja s področja gospodarstva na nove kon-
tekste. Na osnovi jezikoslovnih in zgodovskih znanj proučujemo prisotnost besedja, ki 
je v delu Historical Thesaurus of the Oxford English Dictionary umeščeno v sekcijo 
“trgovina in finance”, v gospodarskih besedilih, vključenih v zbirko Eighteenth Centu-
ry Collections Online. Izposojeno besedje s področja gospodarstva prepoznavamo na 
osnovi njegove hitrosti spreminjanja in s pomočjo multidimenzionalne analize razloču-
jemo štiri leksikalne vidike gospodarskega diskurza, za katere menimo, da se nanašajo 
na (1) javne prihodke in stroške, (2) javni dolg, (3) finančni sistem in (4) zasebna pod-
jetja. Elementi besedja, povezani s temi vidiki, so večinoma latinskega ali francoskega 
374
izvora, številni so novotvorjenke, ki so prvič izpričane v poznem 18. stoletju, kar kaže 
na to, da so takrat na široko prodirale v jezik. Pokazati želimo, da je bila na začetku sto-
letja raba že obstoječih leksikalnih elementov navadno bolj konkrentna in specifična, 
medtem ko so se v abstraknejših in širših kontekstih tovrstni elementi začeli uporabljati 
proti koncu stoletja. To kaže na specializacijo gospodarksega diskurza, ki je povezana 
z nastankom politične ekonomije kot znanstvenega področja.
Ključne besede: korpusno jezikoslovje, zgodovinsko jezikoslovje, multidimenzional-
na analiza, Eighteenth Century Collections Online, proučevanje 18. stoletja