This paper explores the development of Finnish into a standardized language of politics, science and culture in the nineteenth century. For contemporaries, this meant that Finnish could be regarded as a language that supported Finland as a state. We assumed that this expansion of the domains of use of written Finnish would have necessitated the development of more nuanced ways of expressing opinions and attitudes. We studied this by charting the overall frequency of modal expressions as well as the share of epistemic/evidential adverb types among modal adverbs. We found that the share of modal expressions increased in conjunction with the expansion of the Finnish-language press in the 1840s and again in the period after 1880. More importantly, we found that the amount of very frequently used epistemic/evidential adverb types increased in the mid-1800s, meaning that on average, Finnish newspaper writers had more means for nuanced expression in the latter half of the century.
the early modern and modern periods entailed a transformation in linguistic geography
where diglossic systems
Vernacularization can be studied from a number of linguistic and historical
perspectives ranging from changes in the publication landscape and the transformation of how
people valued different languages to the development of the languages themselves. This
paper focuses on the latter perspective by analyzing changes in the Finnish language in
a diachronic corpus consisting of digitized newspapers from 1820–1917. We argue that
as Finnish became the language of politics and cultural criticism, authors needed more
nuanced literary rhetorical conventions to address a new reading public. An increase in
printed material
We propose that the strengthened status of Finnish, and the expansion of its domains of use, is reflected in changes in Finnish modal grammar. Once Finnish became more readily used in public debate, more nuanced and complicated structures emerged to fulfil the newly arisen rhetorical need to evaluate the certainty and reliability of knowledge. In this study, we focus on the emergence, frequency, and distribution of epistemic modal expressions. We claim that these specific linguistic resources, used to align authors’ views with those of their perceived audiences, are robust markers of larger linguistic change that took place when language transcended from mostly agrarian spoken language to a literary and administrative language. 2
Our data set consists of newspapers published in Finnish between the years 1820 and 1917. These have been digitized and made available for data and text mining by the National Library of Finland, with morphological, syntactic and NER annotations provided by FIN-CLARIN. They consist of 5.2B token words and provide a nearly complete record of newspapers and periodicals in the country published in this period. The newspaper corpus as such cannot be regarded as representative of the Finnish language in general, but it is the best historical corpus available, as newspapers covered a wide range of topics and recorded new features in language by publishing everything from poetry to reports on political events and reflections on academic texts.
We study a complex historical process with linguistic markers that are robust to OCR noise and variation typical for historical corpora. Modal adverbs and modal verbs are suitable targets for our purposes because they form a closed set and have limited morphology; furthermore, epistemic and evidential expressions have a direct connection to pragmatic functions in emerging domains of discourse. In this paper, we focus on the 19 most frequent epistemic and evidential modal adverbs, including the Finnish equivalents of e.g. oletettavasti (‘presumably’), ehkä (‘maybe’), and varmasti (‘certainly’), as well as four modal verbs – voida (‘can’), pitää (‘have to’), saattaa (‘may’), täytyy (‘must’) – with stable epistemic use in Finnish.
The sources of newly developed expressions fall roughly under three categories: 1) expressions that already existed in the language, but which undergo a semantic or syntactic change; 2) translations and calques from languages that are further ahead in the vernacularization process (in the case of Finnish, most calques are from Swedish) and thus provide models for what kind of expressions are presumably required to fulfil literary, administrative and public functions; 3) previously unused formations based on the language’s own resources with no obvious outside models.
In analyzing the development of modal expressions, a key object of interest besides general frequency patterns is determining the composition of modal expressions as a grammatical subsystem. This requires comparing modal expressions against general tendencies in their wider grammatical categories. Further, as we are interested in the emerging distinctions among these expressions, we look at contexts where several of them are used. These contexts are interesting because the distinctions between two expressions with similar meanings are often evoked by their juxtapositions. An increase in the density of these contexts could thus plausibly point towards the emergence of more fine-grained communicative resources. We anticipated that there would be considerable changes in the ecology of epistemic modal expressions, these changes mainly taking the shape of the specialization of functions for a number of expressions. The most robust signals for these changes are perhaps the expressions’ relative frequencies, the mapping of which is a relatively trivial task. These markers will be the focus of this paper. 3
In our first experiment we analyzed the shares of modal expressions in individual paragraphs. The emergence of wider contexts of use for written discourse not only required new grammatical constructions functioning as rhetorical devices, but also the emergence of more definitive functional differences between these constructions. We thus presumed that semantic and pragmatic distinctions between the different modal expressions became entrenched during the period studied. In order to quantitatively analyze contexts where these distinctions could plausibly surface, we looked at paragraphs where at least two different modal expressions were used. From these, we calculated the number of modal expressions relative to the full word count of the paragraph.
Figure 1 presents the development of relative frequencies of modal verbs (blue) and
modal adverbs (red) in time bins of ten years, overlapping by five years. The thick lines
mark the mean frequency of each group, and the coloured area the 90% confidence
band. The first data point is thus 1820–1829, the second 1825–1834 and so on. The
temporal dynamic fits well with the development of publication activity in Finland.
While there is an overall rising trend in both groups (a linear regression model shows a
positive slope), there are two periods of rapid rise in the density of modal expressions.
The first one (grey shadow on the left) corresponds with early enthusiasm for
developing the Finnish-language press. This phase is cut short by the strict and wide censorship
legislation imposed by the Russian authorities in the 1850s, which caused stagnation in
the diversification of the press even after the legislation itself was revoked. The second
period (grey shadow on the right) matches the full, wide-scale expansion of the use of
Finnish as a vehicle of written communication in a rapidly modernizing society
In the second experiment (Fig. 2) we further grouped the studied modal adverbs into four categories (low, mid-low, mid-high, high) according to their frequency. Because the frequencies are subject to changes in corpus composition for each time segment, the categories were based on frequencies of all adverbs, a much larger and more varied group than its modal subgroup.
From the frequency distributions of all adverbs we obtained frequency categories for top 5% (“high”), top 5–50 % (“mid-high”), bottom 5–50 % (“mid-low”) and bottom 5% (“low”). The darker the colour, the more epistemic/evidential adverb types there are in that frequency category. In the 1820s, there were hardly any highly frequent modal adverbs. By the 1860s, the selection of the most often used adverbs had grown (Fig. 2). They became a more prominent linguistic feature, manifesting growth in the expressive potential of Finnish when conveying (un)certainties and stances towards reported events. Contrary to the previous analysis, the increase in epistemic/evidential adverb types in the highest frequency category suggests a permanent change in language use already from the mid-1850s onward. This development seems to be to a lesser degree associated with the material development of newspapers. Overall, the analyses nonetheless point towards a similar development. 4
Our study traces historical changes in linguistic features and relates them to the development of the public sphere in Finland. Detailed analysis, based on large-scale historical corpora, of when modal expressions have come to be used in written Finnish is a major contribution to the study of the Finnish language.
We predicted an increase and diversification in the use of modal expressions. This seems to hold on a general level. However, the development was less straightforward than expected due to political circumstances. The censorship decree of 1850 hampered reporting on politics and culture. This also reversed some of the earlier diversification in the use of modal expressions. Tighter censorship in the 1890s and the abolishment of censorship in 1905 did not have similar effects. By this time written Finnish was established enough to retain these expressions.
We found that changes in the relative frequencies of a selection of the modal
expressions increase in conjunction with key turning points in the Finnish press. However, a
full analysis, including multivariate statistical tests, is still needed for making this
argument in full. This study forms a segment of a more ambitious research programme
where we will track modal expressions’ grammatical, contextual and semantic features
simultaneously and in correspondence to each other. This requires further comparison
and curation of dedicated sub corpora for analysis. The overall approach will be akin
to behaviour profile analysis, where occurrences of the studied linguistic items are
examined across a wide range of variables and then subjected to scrutiny by univariate,
bivariate and multivariate statistical tests
We believe that our observations on the Finnish case, which is rather straightforward with a relatively rapid vernacularization process, may lead to discovering similar trends in other languages as well. If the growth of modal expressions is a feature that reflects the writers’ increasing need to take into account an abstract notion of the general public, we should be able to detect this also in other languages with heavy expansion in the nineteenth century, such as Estonian, Latvian, Lithuanian and Norwegian nynorsk.