«Lexical richness in modern women writers: Evidence from the Corpus of History English Texts»

Isabel Moskowich
«Lexical richness in modern women writers: Evidence from the Corpus of History English Texts»
Revista Canaria de Estudios Ingleses, n.º 72 (2016)

Revista Canaria de Estudios Ingleses | Universidad de La Laguna | San Cristóbal de La Laguna (Santa Cruz de Tenerife) | ESPAÑA


Extracto de apartados en páginas 111, 115-116 y 126-127 de la publicación en PDF. Véanse las referencias en la publicación original del texto.

«Abstract

»This paper addresses the issue of lexical density and its popularity after the arrival of corpus linguistics and its methodology. In fact, this is now one of the most frequently used descriptive tools in the analysis of register and genre. Researchers have often trusted lexical density as it is quantifiable and measurable by applying a formula and this has made its use very popular both for scrutinising grammatical and lexical forms and their frequencies. Lexical richness is a related concept although it does not refer exactly to the same, This paper aims to examine lexical richness, understood as the degree of variety of terms used in texts written by women during the eighteenth and nineteenth centuries. To this end, I will analyse samples drawn from the Corpus of English History Texts (CHET) to see whether the communicative format (genre) of the sample has any influence on vocabulary in a discipline with discursive patterns that were not probably as standardised as those of other fields of knowledge.

»Keywords: Lexical richness, late modern English, scientific writing, Coruña Corpus, women authors.

[...]

»Lexical richness and methodology

»While late Modern English is a period of lexical innovation often fostered by social and technological developments, my concern here is not with such innovation but with lexical variety or richness. My initial hypothesis is that texts contain fewer different types as they become more specialised, since some terms are always preferred as being appropriate to refer to particular extra-linguistic entities in specific domains. That is, lexical richness decreases over the course of time as vocabulary becomes more discipline-related.

»Although I agree with Smitterberg and Kytö (129) that genre is an indispensable parameter in historical linguistics, its role as a limiting factor will not be explored in the present study since my sample is a small one, certainly not sufficient to produce any definitive results in this respect. Besides, genre may not be a constraint if we consider that genres have not always been well delimited, as shown for the genre “letter” by Kytö and Romaine (213), where they found that substantial differences could be observed within the genre as a result of revisions made by authors: “professional letters can be highly informational and are often written and revised with care. Thus, they often show a greater degree of lexical variety and informational density compared with personal letters.” We also know that the vocabulary of English increased from the eighteenth century onwards, especially with the introduction of new terms adopted from other languages arising from contact through colonisation and commerce.

»Such an increase did not stop in the following hundred years (Görlach 1999, 93), and indeed, “many of the new words in the OED from the Late Modern English period reflect recent advances in technology” (Tieken-Boon van Ostade 65). It is therefore my intention to look at whether specialised texts reflect the growth in vocabulary in general or, rather, if there is some kind of stagnation due to the adoption of history-related terminology imposed by genre. This will be done by assessing texts in search of the true extent of their lexical variety, density or richness.

»Some discussion seems in order here, since, as mentioned above, lexical richness and lexical density are often used as interchangeable terms. The latter is very clearly defined in computational linguistics as the estimated measure of content per units, either functional or lexical, and is calculated according to a formula the results of which are used in discourse analysis as a parameter varying across registers and genres. In general terms, the calculation establishes the proportion of lexical (nouns, adjectives, verbs, adverbs) tokens in relation to the total number of tokens in the text.

»However, while this measure provides a good portrait of informational density (that is, the proportion of lexical and functional words in the text), it does not really express the idea of variety as such, that is, the many different terms that can be found in a particular sample. For this, Peirce’s 1906 type-token distinction, in which a descriptive class is distinguished from the elements that instantiate that class, is fundamental. In terms of linguistic variety, types (the class or lexeme) and not only tokens (each use of a form) must be taken into account. Content analysis, in Krippendorff’s (24-31) concept of the term, may also be said to be at stake in dealing with the analysis of vocabulary in its context, if we assume that the words mentioned most often (that is, with the greatest number of tokens) are those reflecting important concerns in the texts.

»There are also other ways to measure lexical richness, such as Brunet’s Index (W), Honore’s Statistic (R) and Type-Token Ratio (TTR). The latter has been seen to be unsatisfying in some studies on child language (Richards 1987) and has been said not to be directly related to what one would consider a rich language, yet it may be useful when dealing with samples of a similar size as the ones under study here, and hence it will be used for this study. However, TTR analysis will be complemented with typical frequencies using normalised figures, as well as with a detailed account of each text, a more typical approach in microscopic analyses.

[...]

»Conclusions

»According to the analysis of the Oxford English Corpus carried out jointly by Oxford Online and the Oxford English Dictionary, the is the most common word in the English language, and the same is true of my material. All the texts analysed, not surprisingly, are characterised by having functional words as the most frequently found tokens, the being the first in all cases. However, there are some differences among them. When counting content words in decreasing order of frequency we can see that they appear with different distributions, and that these may be directly related to subjectmatter as well as to genre.

»As Tse and Hyland (177) claim, the use of community discourses helps speakers become members of social groups, defining them in relation to others. They go on to argue that institutional contexts privilege certain ways of making meanings, and it is in this sense that I have used the words “specialised vocabulary” and “domain-specific vocabulary.” In a way, using a particular set of words helps create some kind of professional identity. But the women whose works I have studied here, although discipline-insiders, operated outside institutionalised circles due to the circumstances in which they lived in the late Modern English period. Tse and Hyland also claim that discipline is an important source of variation (179) and this is why I have chosen to study one single discipline and to explore what happens within it.

»Language choice seems to be heavily influenced by discipline and subject matter much more than by gender, and thus I have chosen not to compare gender-related language but language, vocabulary in particular, and to do so within the discipline of history at its very beginnings as an independent field of knowledge, after the emergence of Empiricism, once science had stopped being a totum revolutum and when different branches with their own rules began to appear.

»My analysis seems to indicate that the language of these female authors is influenced not only by the genres they are using, but that these are chosen precisely because of the writers’ intended readerships. Following the tendency noted by Jücker and Kopaczyk (2), I believe that the use of language by authors must not be considered in isolation but rather through considering it within its community and social-cultural context.

»In this sense, the authors here may have been constrained quite severely when writing by the thought of whom they were addressing. In other words, the choice of vocabulary is conditioned by subjectmatter and genre and the latter, in turn, may be influenced by the potential or expected readership, since language use depends on the social and historical context of speakers constituting a particular epistemic community.»