Action nominalizations in early modern scientific english

Resumo

The present dissertation, Action nominalizations in Early Modern scientific English, was conceived as a contribution to the literature on nominalizations. Its point of departure was the assertion that action nominalizations are the result of a word-formation process which aims at filling gaps in the vocabulary of a particular language, English in this case. Action nominalizations are clear cases of grammatical metaphor (Halliday 2004 [1985]), since they are nouns, but they refer to actions as verbs do. For this reason, attention is given to the evolution and use of action nominalizations in the Early Modern English period (henceforth EModE), the time which sees the greatest increase of vocabulary in the history of the English language. Given that nouns prototypically refer to objects rather than actions, the question arises as to how they behave when they denote actions, and what the consequences of this use are. The specific object of analysis are nominalizations involving the suffixes ¿ing, ¿(at)ion, ¿ment, ¿ance, ¿age, ¿ure, and ¿al as represented in the scientific texts of The Penn-Helsinki Parsed Corpus of Early Modern English (PPCEME; Kroch, Santorini and Delfs 2004), and in a selection of texts from the collection Early Modern English Medical Texts (EMEMT; Taavitsainen et al. 2010). The analysis intends to pay close attention to aspects such as: a. the use, distribution and productivity of the above mentioned suffixes; b. the syntactic function of the action noun in the sentence and the number and type of constituents that collocate with action nouns; c. the structural and sociolinguistic variables responsible for changes in frequency and usage, if any; d. the way in which these nominals were acquired into the English language. As noted above, the register chosen for the analysis is scientific writing. This is justified by the fact that nominalizations have traditionally been considered as one of the characteristics distinctive of scientific English. Furthermore, the language of science in the EModE period is of special linguistic interest because it is at that time when scientific writing start to be produced mostly in English, rather than in Latin; until then, as is well known, English is not seen as `eloquent¿ enough for the transmission of knowledge. The study might therefore allow for the detection of interesting changes in the stylistic resources deployed by that register in the course of time. In order to achieve these aims this dissertation has been divided into seven chapters, which will be summarized in what follows. Chapter 1 (¿The concept of nominalization¿) determines the object of study, that is, action nominalizations, and provides a definition of the term. It also deals with the problematic issue of their categorization, given that they are not prototypical nouns in that they do not refer to objects. Furthermore, it also offers a summary of the main approaches to nominalization found in the literature, paying special attention to the systematic relations between verbs and their corresponding nominalizations as in Lees¿ (1968 [1960]). Chomsky (1964) describes nominalizations as transformations deriving from base structures, whereas linguists such as Grimshaw (1990) have looked at the argument structure of nominalization and concluded that only complex event nominals such as construction have argument structure, as do verbs. Chomsky¿s (1970) lexicalist theory is discussed. Here, Chomsky questions his own transformational theory, proposing that the structural differences between derived nominals (refusal) and gerundives (refusing) make it impossible for them to have the same origin. Chomsky develops the lexicalist theory by incorporating syntactic features to the theory of grammar, this being used to explain deverbal nominalizations. Attention is also paid to the idea that nominalizations form a squish in which they go from the more nominal to the more verbal kinds of nominalizations (cf. Ross 1973, 2004; Mackenzie 1996); nominalizations as instances of grammatical metaphor, as proposed by Halliday (2004 [1985]), Halliday and Martin (1993) and Banks (2008); and studies in which nominalizations are analyzed in context, that is, their use is assessed in particular kinds of texts (see Albentosa-Hernández and Moya-Guijarro [2001]; Andersen [2007], among others). Chapter 2 (¿¿Ing nominals: origin and development¿) is devoted to ¿ing nominals (building, refusing, etc.), since they are by far the most controversial among the nominalizations analyzed in this study. Thus, this chapter discusses the origins and development of ¿ing nominals, focusing on their syntactic behaviour during the EModE period. As clarified in this chapter, back in OE the ¿ing suffix was used to create action nouns from verbal bases. However, over the course of time these formations have been acquiring verbal syntactic features such as the ability to take a subject in a case other than the genitive (Jane arriving late was surprising) and the ability to take an object without the preposition of (by eating apples), among others (see Jespersen 1940 [1909] Vol. V: 89-90, Fanego 1996: 98), a fact that has led to the coexistence of the two kinds of ¿ing nominals above mentioned. As for their chronology, here it is made clear, although not discussed exhaustively, that the English (verbal) gerund construction becomes available in ME. This chapter also goes on to describe the structural instability of ¿ing formations during the EModE period; at this point, then, not only did nominal (the reading of the poem) and verbal (reading the poem) ¿ing forms coexist, but also mixed forms of the kinds (his reading the poem) and (the reading the poem). These mixed ¿ing formations are the result of a non-uniform verbalization process, which takes place earlier in more oral/informal genres and in nominalizations having only post-head dependents (by reading the poem), and only later in those showing both pre- and post-head dependents by a process of extension (Harris and Campbell 1995). Since post-head dependents start to be verbal, whereas pre-head ones remain nominal, mixed forms appear and coexist for a while with both nominal and verbal ones. These mixed gerunds are relatively common since they can be used in slots where verbal formations were not yet accepted (Fanego 1996). Chapter 3 (¿Nominal complementation and argument structure¿) discusses the issues of argument structure and nominal complementation, and reviews the literature on the topic. In addition, it gives a detailed overview of the nominal dependents that nominalizations can take. In this vein, it discusses the similarities between the items that co-occur with nominalizations and those that co-occur with their base verbs. Here the idea of valency reduction (Mackenzie 1985: 32) is also introduced. Finally, the topic of argument structure in nominals is presented, listing the main views in support of it (Vendler 1968; Zubizarreta 1987 and Grimshaw 1990) and those against it (Anderson 1983; Higginbotham 1983 and Dowty 1989). However, the definition of argument structure is fully developed later. The main dependents of action nominalizations are also analyzed. As a means of classifying them, they have been divided into pre-head dependents and post-head dependents following Fanego¿s (1996) model. The items found in pre-head position are usually determiners, possessives (not only pronouns but also genitive NPs), adjectives, nouns and adverbs. As for post-head dependents, the literature reveals that they are usually of-PPs and by-phrases, and to a lesser extent other PPs as well as NPs and sentential complements. The dependents of nominalizations have a parallel role to arguments in sentences, but the number of arguments appearing in nominalizations is usually considerably lower than in their non-nominalized counterparts, the process of nominalization usually implying a valency reduction. This assertion is subsequently tested empirically in Chapter 6. Chapter 4 (¿Rhetoric and the world of science in the EModE period¿) deals with the rhetoric of scientific writing and examines the status of science during the EModE period since, as is well known (Atkinson 1999: xviii, Pahta and Taavitsainen 2011: 1-2), the influence of the sociohistorical context has been proved to be very important when dealing with linguistic variation. Scientific writing is described as a heterogeneous genre, but its general features are nevertheless listed, paying special attention to those related to the use of nominalizations. It is claimed that the use of nominalizations helps to achieve the objectivity and impersonality required in scientific texts as well as to describe actions that took place in the scientific experiments. Hence, nominalizations are useful devices for the redistribution of information in a clause and to help in the ¿step by step¿ description of scientific processes. They are also useful as cohesive devices. The role of the Royal Society and the printing press in the homogenization of scientific writing and in the spread of science to a wider community, including the middle classes, is also analyzed. The changes in the language of science, from the use of Latin to the vernacularization of science are also described. It is stated that Latin is the lingua franca of science during the Middle English period. Latin is at this time the barrier that separated learned from lay men as well as separating science from popular tradition. However, in the EModE period the situation changes and scientists come to associate Latin with antiquity and traditional notions of science. Despite being the key to accessing a vast cultural heritage, Latin loses ground in favour of the vernacular, which is increasingly used in even the more prestigious registers. The different problems in the vernacularization of science are also analyzed. As Jones (1953) and Gotti (2006) have both discussed extensively, the greatest problem faced by the vernacular is a lack of eloquence and of technical terms, which renders it unfit for the scientific register. As a way of expanding vocabulary, two main trends emerge: creating a new word or giving a specialized meaning to an existing one, and borrowing words from other languages, especially from the prestigious classical ones. However, the extensive borrowing that takes place during this period is strongly criticized by some authors at the time, who consider these new words difficult to understand. Medical texts and the categories of writing therein are also described. The field of medicine in EModE is more heterogeneous and less clearly defined than nowadays. However, the different kinds of medical writings are perfectly delimited, and range from most popular to most academic. In Chapter 6 of this dissertation, remedy books, surgical treatises and academic treatises are analyzed in detail, with findings showing differences in the use, origin and frequency of nominalizations in these three types of texts. Remedy books belong to the vernacular tradition and are intended for a wider audience, read not only by doctors but also by the middle classes. Surgical treatises, in turn, are learned texts compiled by doctors and used mainly by surgeons and trainee surgeons. However, since their use is mainly practical, they are not considered to be learned works. By contrast, academic treatises are used to teach at universities, and sometimes deal with very specific topics; thus, their audience is learned and specialized. Chapter 5 (¿Corpus and methodology¿) contains a description of the corpora used and the methodology employed. This chapter tries to throw some light on the process of decision-taking. In other words, the reasons behind the selection of the corpora, and the specific suffixes examined are all clarified here. It also describes the classification of nominals which are used in the analysis in Chapter 6. These are classified according to three different parameters, namely, constituents of the NP (i.e. the nominal head as the sole constituent, only pre-head dependents, only post-head dependents, and both pre- and post-head dependents), their function in the superordinate sentence (i.e. subjects, objects, predicatives, complements of a preposition, supplements or modifiers, and absolute position), and their internal syntax (i.e. nominal, verbal or mixed). An account is also given on a number of problematic issues encountered when classifiying the data. Chapter 6 (¿Findings¿) is a corpus-based analysis concerned with the use and development of nominalizations in Early Modern scientific English. Chapter 6 reports on the empirical work that is intended to complement Chapter 3. It offers a thorough description of the use of ¿ing and Romance nominalizations in Early Modern scientific writing. Section 6.1 gives an overview of the evolution and use of nominalizations at the time. The corpus study indicates that both ¿ing and Romance nominalizations are on the rise in scientific texts. This increase in their frequency of use is more remarkable in the case of Romance nominalizations, which is explained as a result of the need for the vernacular to adapt for use in the scientific register. For example, the lack of terms related to art was overcome by the use of these formations, especially so in the case of Romance nominalizations, which are adopted from prestigious languages at the time such as French and Latin. The examination of structural variability in nominals reveals that the increase in the number of nominalizations does not apply to all the structural types in the same way. Thus, in the case of ¿ing nominalizations, the decrease of nominal formations in any type of phrase runs in parallel with an increase of verbal ¿ing forms, especially those having only post-head dependents. However, Romance nominalizations are observed to behave in a different manner, and are preferred in structures having both pre- and post-head dependents. Therefore, the analysis points towards a possible specialization of the ¿ing suffix to be used in verbal nominalizations. The gap left by nominal ¿ing nominalizations is filled by Romance nominals. Taking into account the differences in the kind of phrase of both ¿ing and Romance formations, it seems important to consider also the constituents of these phrases. As with the type of phrase, the kind of pre-head dependents taken by ¿ing formations differ considerably depending on the nature of the nominalization in question. Hence, nominal ¿ing formations are usually found with determiners, and adjectives and possessives to a lesser extent. In contrast, verbal ¿ing formations are hardly ever found with pre-head dependents. Romance nominalizations in turn follow the same trend as nominal ¿ing forms, with determiners, adjectives and possessives being the preferred options. As for post-head dependents, once again we see that verbal ¿ing nominalizations behave differently from nominal ¿ing and Romance nominalizations. Whereas the verbal ¿ing forms are generally followed by NPs, nominal ¿ing and Romance nominalizations are followed by of-PP[object]s and of-PP[subject]s. The kind of post-head dependents used by nominals appeared to be related to their nature. Thus, nominal ¿ing and Romance nominalizations are nouns properly speaking and therefore cannot take NPs as post-head dependents, whereas verbal ¿ing formations can, being strongly reminiscent of the pattern of a verb followed by a DO. There were also mixed ¿ing forms, having the pre-head dependents used with nouns, but NPs as their post-head dependents. This wide range of possibilities accounts for the structural complexity of nominalizations at the time. As for their syntactic behaviour, nominalizations are preferred as complements of a preposition, especially verbal ¿ing nominalizations. The productivity of the suffixes is also analyzed according to variables such as their frequency and transparency. Results show that ¿ing is the most frequent suffix in the nominalizations found in the corpora. It is combined with both native (working, seeking) and non-native bases, especially Romance ones, creating hybrids such as crossing and evaporating, but also with bases from ON, such as mistaking. This indicates that the suffix is transparent, in the sense that speakers used it to create new action nouns. As for Romance suffixes, their frequency increases enormously during the EModE period, the most frequent being ¿(at)ion. However, this suffix always appears attached to Romance bases (attraction, suppuration), which makes it difficult to determine whether these formations are created in English following Latin patterns or if they have been borrowed (Section 6.2 tried to shed some light on this issue). The suffix ¿al is not found in nominalizations from either corpora, and the other Romance suffixes are considered marginal if compared to ¿(at)ion due to their much lower frequencies. Of particular interest is that all hybrids found are formed with these low-frequency suffixes (ailment, hindrance, tarriance and tillage). The dramatic increase of Romance nominalizations during the EModE period, as well as the difficulties when trying to distinguish the method of acquisition (Nevalainen 1999: 397), justify the necessity of Section 6.2. This section seeks to analyze how these formations became part of the English language, as well as to clarify the factors that might have favoured them becoming part of the vernacular. With these aims in mind, variables such as their bases and the chronology of their formation are taken into account. Data from this section has clarified that, contrary to what might have been expected, most of these Romance formations are already part of the English vocabulary in ME. It must be conceded that some of them are attested in E1 and E2, coinciding with the peak in borrowing as dated by Görlach (1991: 137) and Barber (1997 [1976]: 222). However, it is clear that the majority of these formations are adopted into English before the great lexical expansion that takes place in the EModE period (Finkenstadt et al. 1970). It is also established that most of the terms are borrowings from Latin and French. This finding is in line with those of Nevalainen (1999: 351) and Görlach (1991: 155). The scarcity of English coinages is explained by the fact that the range of environments in which a Romance suffix is allowed is somehow more restricted than those of the native ¿ing. Finally, it can be claimed that there is no clear, demonstrable influence of translation in the borrowing of Romance nominalizations at the time. A possible explanation for the similar number of Romance formations in both translations and English texts is that authors writing English surgical treatises are learned and thus would have a good command of classical languages. Thus, they use Latin texts as a model and fill any gaps in the vocabulary of the vernacular by using these Romances terms. It is widely acknowledged that scientific writing is heterogeneous due to the wide variety of topics and audiences treated (Halliday and Martin 1993: 54). Table 8 in Section 6.1.1 shows that there are differences in the frequency of nominalizations according to the nature of texts. Therefore, a closer analysis of the possible variation in the use of nominalizations according to variables such as audience and text category is required. Section 6.3 serves this purpose. Here three categories of medical texts are chosen, remedy books, surgical treatises and academic treatises. On the popular-learned continuum, remedy books are the most popular text type and academic treatises the most learned. All the texts analyzed in this section were drawn from the EMEMT. As expected, findings show that there are significant differences in the frequency and origin of the nominalizations used in these categories of medical writing. Thus, remedy books show the lowest number of nominalizations and academic treatises the highest. It is also noticeable that Romance nominalizations are used far more in both surgical and academic treatises than in remedy books. This can be explained in terms of audience and writing tradition. Remedy books have a popular, literate audience for whom specialized terminology such as Romance nominalizations might prove to be obscure. However, the learned audiences of both surgical and academic treatises are far more likely to have a sound understanding of these difficult words. Furthermore, whereas remedy books have been written in the vernacular since OE times, surgical and academic treatises are originally written in Latin and, once they start to be composed in the vernacular, authors persist in using Latin texts as models, favouring the use of Romance formations. Finally, Chapter 7 (¿Conclusions and suggestions for further research¿) offers a summary of the investigation as well as the main conclusions reached. Suggestions for further research are also included. This dissertation has offered a thorough analysis of action nominalizations in EModE, focusing on their development and use in scientific texts. As for their development, ¿ing nominalizations evolve to a more verbalized structural pattern, in which the preferred structure is that of ¿ing forms having only post-head dependents (NPs). However, this is a gradual process and at the time nominal, mixed and verbal ¿ing formations co-exist. The gap left by the increasingly infrequent nominal ¿ing nominalizations is filled by Romance formations. As regards their use, it has been shown that this differed according to variables such as audience and text category. As far as their function is concerned, it has been shown that nominalizations begin to be used as grammatical metaphors in the packaging and distribution of information, in that they were very useful for the agglutinative style of scientific writing. However, apart from this primary structural function, nominalizations develop a second one: they become markers of scientific style. REFERENCES: Albentosa-Hernández, José Ignacio and Arsenio Jesús Moya-Guijarro. 2001. ¿La reducción del grado de transitividad de la oración en el discurso científico en lengua inglesa¿. Revista Española de Lingüística 30, pp. 445-468. Andersen, Øivin. 2007. ¿Deverbal nouns, lexicalization and syntactic change¿. Nordic Journal of Linguistics 30, pp. 55-86. Anderson, Mona Katherine Johnson. 1983. ¿Prenominal Genitive NPs¿. The Linguistic Review 3, pp. 1-24. Atkinson, Dwight. 1999. Scientific Discourse in Sociohistorical Context. The Philosophical Transactions of the Royal Society of London, 1675-1975. Mahwah, New Jersey: Lawrence Erlbaum. Banks, David. 2008. The Development of Scientific Writing. Linguistic Features and Historical Context. London: Equinox. Barber, Charles. 1997 [1976]. Early Modern English. Edinburgh: Edinburgh University Press. Chomsky, Noam. 1964. ¿A transformational approach to syntax¿. In The Structure of Language. Readings in the Philosophy of Language, pp. 211-245. Ed. by Jerry A. Fodor and Jerrold J. Katz. Englewood Cliffs, New Jersey: Prentice-Hall, Inc. -----------. 1970. ¿Remarks on nominalization¿. In Readings in English Transformational Grammar, pp. 184-221. Ed. by Roderick A. Jacobs and Peter S. Rosenbaum. Waltham, Massachusetts, Toronto, London: Ginn and Company. Dalton-Puffer, Christiane. 1996. The French Influence on Middle English Morphology. A Corpus-based Study of Derivation. Berlin and New York: Mouton de Gruyter. Dowty, David R. 1989. ¿On the semantic content of the notion `Thematic Role¿¿. In Properties, Types and Meaning, vol. II: Semantic Issues, pp. 69-129. Ed. by Gennaro Chierchia, Barbara H. Partee, and Raymond Turner. Dordrecht, Boston and London: Kluwer Academic Publishers. Fanego, Teresa. 1996. ¿The gerund in Early Modern English: Evidence from the Helsinki Corpus¿. Folia Linguistica Historica 17, pp. 97-152. Finkenstaedt Thomas, Ernst Leisi and Dieter Wolff. 1970. A Chronological English Dictionary. Heidelberg: Winter. Görlach, Manfred. 1991. Introduction to Early Modern English. Cambridge: Cambridge University Press. Gotti, Maurizio. 2006. ¿Medical discourse: Developments, 16th and 17th centuries¿. The Encyclopedia of Language and Linguistics, vol. VII, pp: 674-681. 2nd edition. Ed. by Keith Brown. Amsterdam, Boston, Heidelberg, London, New York, Oxford, Paris, San Diego, San Francisco, Singapore, Sidney and Tokyo: Elsevier. Grimshaw, Jane. 1990. Argument Structure. Cambridge, Massachusetts, London, England: The MIT Press. Halliday, Michael A. K. 2004 [1985]. An Introduction to Functional Grammar. 3rd edition (revised by Christian M.I.M. Matthiesen). London and New York: Oxford University Press. Halliday, Michael A. K. and James R. Martin. 1993. Writing Science. London: The Falmer Press. Harris, Alice C. and Lyle Campbell. 1995. Historical Syntax in Cross-linguistic Perspective. Cambridge: Cambridge University Press. Higginbotham, James. 1983. ¿Logical form, binding and nominals¿. Linguistic Inquiry 14, pp. 395-420. Jespersen, Otto. 1940 [1909]. A Modern English Grammar on Historical Principles, vol. V: Syntax. London: George Allen and Unwin. Jones, Richard Foster. 1953. The Triumph of the English Language. A Survey of Opinions Concerning the Vernacular from the Introduction of Printing to the Restoration. Stanford, California: Stanford University Press. Kroch, Anthony, Beatrice Santorini, and Lauren Delfs. 2004. The Penn-Helsinki Parsed Corpus of Early Modern English (PPCEME). Department of Linguistics, University of Pennsylvania. CD-ROM, 1st edition, (http://www.ling.upenn.edu/hist-corpora/). Lees, Robert B. 1968 [1960]. The Grammar of English Nominalizations. The Hague: Mouton. [Originally published as Publication 12, Bloomington: Indiana University Research Center in Anthropology, Folklore and Linguistics.] Mackenzie, J. Lachlan. 1985. ¿Nominalization and valency reduction¿. In Predicates and Terms in Functional Grammar, pp. 29-48. Ed. by A. Machtelt Bolkestein, Caspar de Groot and J. Lachlan Mackenzie. Dordrecht: Foris Publications Holland. Nevalainen, Terttu. 1999. ¿Early Modern English lexis and semantics¿. The Cambridge History of the English Language, vol. III: 1476-1776, pp. 332-458. Ed. by Roger Lass. Cambridge: Cambridge University Press. Pahta, Päivi and Irma Taavitsainen. 2011. ¿An interdisciplinary approach to medical writing in Early Modern English¿. In Medical Writing in Early Modern English, pp. 1-8. Ed. by Irma Taavitsainen and Päivi Pahta. Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Delhi, Dubai, Tokyo and Mexico City: Cambridge University Press. Taavitsainen Irma, Päivi Pahta, Turo Hiltunen, Martti Mäkinen, Ville Marttila, Maura Ratia, Carla Suhr and Jukka Tyrkkö (eds.). 2010. Early Modern English Medical Texts. CD-ROM. Amsterdam: John Benjamins Publishing Company. Ross, John Robert. 1973. ¿A fake NP squish¿. In New Ways of Analyzing Variation in English. Ed. by Charles-James N. Bailey and Roger W. Shuy. Washington D.C.: Georgetown University Press. -----------. 2004. ¿Nouniness¿. In Fuzzy Grammar. A Reader, pp. 351-422. Ed. by Baas Arts, David Denison, Evelien Kezer and Gergana Popova. Oxford: Oxford University Press. Vendler, Zeno. 1968. Adjectives and Nominalizations. The Hague, Paris: Mouton. Zubizarreta, Maria Luisa. 1987. Levels of Representation in the Lexicon and in the Syntax. Dordrecht, Holland: Foris Publications.