Middle English, London, Modern English, van der, gerunds, Cambridge University Press, English grammar, De Smet, Amsterdam & Philadelphia, John Benjamins, English Language and Linguistics, Old English, Visser, English Language, American English, history of the English language, verbal constructions, English Studies, Oxford University Press, Teresa, gerunds and infinitives, Katholieke Universiteit Leuven, historical syntax, language standardization, Chaucer Troilus and Criseyde, complements, Joseph Andrews, Van der Wurff, subordinate clauses, Kari E. Haugland, Sundby, Bertil, Robert Dodsley, Tieken-Boon van Ostade, Short introduction to English grammar, Wim van der
93 The role of language standardization in the loss of hybrid gerunds in Modern English Teresa Fanego University of Santiago de Compostela1 1. Introduction Sentential complements, i.e. subordinate clauses functioning as an argument of a higher predicate, as in `I think that she made a mistake' or `To refuse her request would be unthinkable', have been a prolific area of research for many years among English historical linguists; cf., among others, Warner (1982); Fischer (1988, 1989, 1995); Fanego (1990, 1992, 1996a/b/c, 1997, 1998, 2004a/b, 2006); Rohdenburg (1995, 1996); Los (1998, 2005); Rudanko (1998, 1999, 2000); Miller (2002); Wagner (2002); Mair (2002, forthcoming); Vosberg (2003, 2004); or, more recently, De Smet (2004, 2005); and De Smet & Cuykens (2005). While the focus of these studies has been chiefly on the internal factors responsible for the important changes affecting the English complement system since Old English times,2 the present analysis will be concerned with the role of extra-systemic factors, and more specifically of language standardization, in the loss of the type of gerundive clause exemplified in (1).3 1 I am grateful to the Autonomus Government of Galicia (grants nos. PGIDIT02PXIC20402PN and PGIDIT05PXIC20401PN) and the Spanish Ministry of Education
and Science (grants nos. BFF20012914 and HUM200400940) for generous financial support. 2 Though there is no complete agreement among scholars as to when exactly one period in the history of English ends and another begins, I will here adopt the more or less traditional divisions of Old English (up to about 1100), Middle English (11001500), Early Modern English (EModE; 15001700), Late Modern English
94 TERESA FANEGO (1) believing that if he were refused there, he should never get an Opportunity of applying to the young Lady, he chose to make the first Declaration to herself; the gaining her Affections being the material Point, he considered all others of little Consequence. (COLMOBAENG 1725 Haywood Fatal Secret, 217) The paper is organized as follows. Section 2 gives an overview of English sentential complements and summarizes their development since Old English times, with particular reference to the rise of gerundives. Section 3 reviews the proposals put forward in the literature to account for the decline of the gerundive construction mentioned above. Section 4 argues, on the basis of corpus data, for an alternative explanation. 2. The development of English sentential complements: an overview If we leave aside interrogative complements (`I'm wondering why I should go at all', `I doubt whether they knew'), five major types of complement clauses can be distinguished in English, as illustrated in (2)(6): That/zero-declaratives (2) a. It is clear (that) he made a mistake. b. He knows (that) you are here. (LModE; 17001900) and Present-day English, with a further subdivision of Middle English into Early Middle English (11001300) and Late Middle English (13001500). 3 Unless otherwise indicated, all quotations are from the Corpus of Late Modern British and American English Prose (COLMOBAENG, 17001879), a 1,170,000 word electronic database comprising both fiction and nonfiction texts distributed by date in four different subperiods, as follows: 17001726 (BrE only; 200,000 words), 17321757 (200,000 words BrE; 50,000 words AmE), 17611797 (200,000 words BrE; 120,000 words AmE), 18501879 (BrE and AmE; 200,000 words each). For the list of texts and other details, see Fanego (2006).
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 95 Bare infinitives (3) a. All I did was ask a question. b. We saw Kim leave the bank. To-infinitives with and without a subject (4) a. Max wanted to change his name. b. The best plan would be for them to go alone. Ing participle clauses as complements of perception verbs (5) We saw Kim leaving the bank. Gerundive clauses with and without a subject (6) a. Inviting the twins was a bad mistake. b. I resented them / their going without me. Unlike types (2), (3), (4) and (5), which have been on record from Old English times, gerundive clauses emerged in Late Middle English. In this type the ing form is not participial in origin, as in (5), but descends instead from an Old English derivational suffix which could be freely added to verb stems to form abstract nouns of action, as in OE spilling `destruction' (< spillan `destroy') or OE wending `turning' (< wendan `turn'). Following common practice among historians of the English Language
, I will employ the label nominal gerund to refer to this kind of nouns and to their reflexes in Present-day English, as in `the exploring of the mountain took a long time' or `the restructuring of the Government is important'. The labels verbal gerund or gerundive -ing clause will be applied to examples such as (6ab). In Old English and Early Middle English nominal gerunds behaved like any other noun in all relevant respects, and could therefore take nominal dependents of various kinds. The following examples illustrate their use with determiners (the, his) and with of-phrases serving as their notional objects: (7) at the makyng of thys lettyr `at the [moment of] writing this letter / when writing this letter' (14721488 CelyLetters, 94/5 (Tajima 1985:68)) (8) And thus began his loving of Criseyde
96 TERESA FANEGO (c. 1385 Chaucer Troilus and Criseyde, V 1833 (Tajima 1985:70)) (9) Sain Jon was ... bisi In ordaining of priestes, and clerkes, `Saint John was ... busy ordaining priests and clerics,' (c. 1300 (MS a1400) English Metrical Homilies, 112/24 (Tajima 1985:76)) For reasons which I have discussed elsewhere (cf. Fanego 2004a), from Late Middle English onwards nominal gerunds began to acquire verbal properties, a development that has ultimately led to the Present-day English situation, where gerunds have the ability to: (i) govern an object or a predicative complement (e.g. `their following the child into England', `I don't like being ill'); (ii) be modified by adverbs or adverbials restricted to co-occurring only with verbs (e.g. `my quietly leaving before anyone noticed'); (iii) show tense and voice distinctions (e.g. `of having done it', `the necessity of being loved'); (iv) be negated by means of the VP-negating particle not (e.g. `my not leaving'); and (v) take a subject in a case other than the genitive (e.g. `I resented them going without me'). Though, as noted above, the first instances of verbal gerunds can be dated back to Late Middle English, their spread across the grammar of English extended over a period of several centuries, with some subtypes becoming possible much earlier than others. In the early stages, for instance, verbalization was largely restricted to prepositional environments (e.g. `I insisted on wearing a suit'), which were the primary context in which the gerundive was not blocked by the productive toinfinitive, as this was available in Old and Middle English in a variety of clausal functions, but could not occur after prepositions other than to (for discussion see Fanego 1996b, 2004a; Miller 2002:Chapter 11). In addition, for a long time verbal features were found only with those prepositional gerunds that lacked an explicit subject, as in (10), but not with the type exemplified by structures like (8) above (his loving of Criseyde), where the possessive his represents the notional subject of the following ing form.4 4 The development of gerunds introduced by possessives or by the negative determiner no (e.g. `there was no getting a word from her on any other theme') was substantially different (cf. Fanego 1996b, 2004a) from the development of subjectless gerunds and will not be discussed here.
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 97 (10) yn feblyng ?e body with moche fastyng `in weakening the body by too much abstinence' (c. 1303 (MS a1400) Handlyng Synne, HS 408 (Tajima 1985:76)) Gradually, however, verbal gerunds came to be licensed in functions other than that of prepositional complement, so that from the middle of the sixteenth century (see Fanego 1996a:38) we come across scattered instances of gerundives encroaching upon the to-infinitive as the objects of verbs of so-called subject control, that is, verbs with which the covert subject of the complement clause is co-referential with the matrix subject (e.g. `Johni remembered PROi seeing Bill', `Jacki avoided PROi meeting Mary'). From the late seventeenth century occasional examples of verbal gerunds used as subjects or predicatives can also be found, as in these examples retrieved from the Helsinki Corpus of English Text
s (henceforth HC; see Kytц 1996): (11) Slitting the bark is an excellent additional help to most of the foresaid evils, and also for bark-binding, (HC 1699 Langford Plain and Full Instructions to Raise All Sorts of Fruit-Trees, Sample 2, 114) (12) Your Lordship does me too much honour, it was exposing your Person to too much Fatigue and Danger, I protest it was; HC 1697 Vanbrugh The Relapse, I, 59 Roughly about the same time structures such as the following also became common: (13) and though I profess no knowledg of the Law, yet I am sure the regulation of these defects might be easily mended ... But above all, the taking Fish in Spawning time, may be said to be against nature; (HC 1676 Walton Compleat Angler, 213) (14) ... a consideration which I take the liberty to recommend a little to the reader: for however swift his capacity may be, I would not advise him to travel through these pages too fast: for if he does, he may probably miss the seeing some curious productions of nature which will be observed by the slower
98 TERESA FANEGO and more accurate reader. (COLMOBAENG 1742 J. Fielding Joseph Andrews, 066/105-P23) (15) She certainly would soon have broke her Heart, had she known that all this Misery ... was her own Fault; but as she thought it his Inconstancy, to his Generosity, in not telling her the Truth, she owed the avoiding that painful Reflection. (COLMOBAENG 1744 S. Fielding David Simple, 71) (16) ... but of all these [ends] the noblest End is the multiplying children, It is religion to marry for children; (HC 1673 Taylor Sermons, 13) (17) There is one piece of sophistry practiced by both sides, and that is the taking any scandalous story that has been ever whispered or invented of a private man, for a known undoubted truth, and raising suitable speculations about it. (COLMOBAENG 1711 Addison Spectator, 114/040-P06) These are categorially hybrid, in that the gerund governs a direct object, but is preceded by the definite determiner the. In the majority of cases, the occurrence of such hybrids is restricted to those clause slots where the verbal gerund had long been disallowed, that is, they function predominantly as preverbal subjects, as in (1) and (13) above, objects (14)(15) or predicatives (16)(17). There are thus grounds to suggest that, as I have argued in detail elsewhere (Fanego 2004b), the function of the in such instances was not to indicate definite reference, but rather to provide the following verbal gerund with an introductory element of some kind; in other words, the was basically a semantically empty grammatical marker or complementizer serving to license the ing clause at a time when subjectless verbal gerunds in argument positions (i.e. as subjects, objects, or predicatives) were probably not yet fully acceptable. Though the-hybrids were relatively common in the seventeenth and eighteenth centuries, by the end of the nineteenth they had largely gone out of use, as first noted by Visser (1972:§1124) in an oft-quoted passage where he suggested that their decline in frequency in the modern period `seems to be ascribable to [their] being pilloried as solecistic by a number of prescriptive grammarians, who, for once, appear to have been to a large extent successful'. More recently, hybrid gerunds have also been discussed at some length by van der Wurff (1993, 1997),
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 99 whose views on the loss of these nomino-verbal constructions are briefly summarized in the next section.5 3. Van der Wurff's (1993, 1997) account of the loss of hybrid gerunds In two influential papers (1993, 1997) containing many valuable insights, van der Wurff expresses his doubts that Visser's explanation for the decline of mixed gerunds is adequate: a first question that may be asked, he argues, is `why were the grammarians successful in this case, while their objections apparently had so little effect in other cases (for example, their condemnation of had better/had rather ... , and the use of the passive progressive form6)?' (1993:367). Van der Wurff feels, therefore, that another explanation for the changes involving the gerund is called for, even though he seems prepared to admit `that condemnation by the grammarians [of the pattern in (13)(17) above] may have played a role' (1993:368) in its disappearance. His (1993) proposal relies on work by Hoekstra (1986) and argues that the English gerund construction was originally of mixed nominal and verbal character, that is, essentially [+V,+N]. Sometime during the Late Modern English period a split took place, from this mixed type [+V,+N] into two types, one nominal (i.e. [-V,+N]), as in the stealing of the bike was noticed, and the other verbal (i.e. [+V,-N]), as in stealing the bike was noticed, with concomitant loss of hybrid structures like the stealing the bike. The reasons for the change were two independent developments that happened to take place in English during the seventeenth to nineteenth centuries, namely on the one hand the steady growth in the number of action nouns of the type destruction, capture and betrayal, which would exert pressure on the more or less synonymous gerund to become fully nominal, and on the other hand the explo- 5 Hybrid gerunds are also the topic of Moessner (1997). This study, which uses data from Defoe's Moll Flanders (1722; 130,000 words), Goldsmith's The Vicar of Wakefield (1766; 75,690 words) and Dickens' Great Expectations (1861; an extract of 8,000 words), contains many inaccuracies and will not be discussed here. 6 I.e. structures such as `novels were being read to him', which were also stigmatized and heavily condemned in grammar books of the time. Cf. Denison (1993:428ff.).
100 TERESA FANEGO sive increase in the use of the fully verbal progressive, which would provide pressure for the homophonous gerund to become verbal. In his later study, van der Wurff departs from his earlier analysis because he argues, with good reason, that `nominal gerunds have existed in English more or less unchanged from the earliest times' (1997:10), and hence there does not seem to exist a clear justification for analysing them `as nouns in Old English, Middle English and Present-day English, but as hybrid forms in the period 15001900' (1997:10).7 He therefore proposes a different explanation based on Yoon's generative analysis (1996) of Present-day English gerunds, whereby he postulates the existence in earlier English of three different classes of gerundial constructions: one (the hybrid gerund) structurally marked, in that, within his generative framework, its derivation involves attaching several nominalizing and verbalizing zero affixes `to an intermediate category (V', N'), rather than to a head or maximal projection, which are the natural units for syntactic operations' (1997:18); and, secondly, two other classes (the verbal and the nominal gerunds) structurally unmarked, for the contrary reasons. While in his earlier paper van der Wurff had relied exclusively on Visser's collection of data, he now makes use of a small selection of narrative texts dating from 1722 to 1861 and totalling some 70,000 words. The joint evidence from his corpus and from Visser's examples leads him to conclude that during the period under discussion `the vast majority of the gerunds are either verbal or nominal' (1997:10), while `unambiguously hybrid tokens are a very small minority' (1997:12). At this point van der Wurff takes up again his (1993) hypothesis concerning the competition from action nouns, and argues that, as a result of it, the overall frequency of the marked hybrid pattern declined throughout the period, until `in the course of the 19th century [it] dropped below a critical level' (1997:20) and could no longer be acquired by language learners, as the evidence for its existence was not solid enough any more. In his data, van der Wurff also finds that the percentage of nominal gerunds is very low by comparison with that of verbal gerunds 7 As pointed out earlier in this paper, hybrid gerunds become noticeable only from the seventeenth century onwards. When van der Wurff refers to `hybrid gerunds in the period 15001900' he is relying on Visser's dating of the type (1972:§1124), which is much too early, as I have demonstrated elsewhere (see Fanego 1996b: 13334, 1998).
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 101 (these representing 82% of all gerunds in the eighteenth century and 87% in the nineteenth century), yet the nominal gerund survived because it was an unmarked option `in comparison with the mixed type, ensuring acquisition even on the basis of relatively few examples' (1997:20). Van der Wurff's analysis, which is largely endorsed by Denison (1998:27172) and other researchers, contains valuable observations, and also some shortcomings, one such being the nature of the evidence he uses: overall, gerunds are infrequent constructions in English,8 so it is difficult to reach any reliable conclusions on gerundial syntax with a corpus the size of van der Wurff's.9 In view of this, I considered it worthwhile to examine in detail whether Visser's hypothesis regarding the influence of prescriptive grammar on the loss of the-hybrids could be proved correct or not. To this end, I retrieved all the gerunds introduced by the in a large corpus of British English comprising both fiction and nonfiction texts and covering the period 16401879. The results of my search, as displayed in Table 1,10 are discussed in Section 4. A few explanatory comments on the table are given in Table 1. Excluded from the above figures are nominal gerunds such as: `by the heeling to and fro of the ship' (COLMOBAENG 1739 Hales Philosophical Experiments, 6N07 0043/085-P0), where of the ship represents the notional subject of heeling, rather than its object. Only gerunds governing notional objects are relevant to the present discussion, since they alone allow their object to surface either as a NP (the gaining her affections) or an of-phrase (the gaining of her affections). The application of the chi-square test to the data for hybrid and nominal gerunds in the four Late Modern English subperiods yields the 8 This comment does not apply to prepositional gerunds lacking an overt subject (e.g. by/with/on asking it), which are indeed very frequent in English; cf. Fanego (1996b). 9 Another serious objection to van der Wurff's account is of course that it cannot explain the survival into Present-day English of hybrid structures introduced by possessives (`I insisted on his wearing a suit') or by the negative determiner no (`there was no getting a word from her on any other theme'); see footnote 4 above. 10 The data in the first row are drawn from a sample of the Helsinki Corpus comprising ten different genres, namely Diaries, Private Letters, Fiction, Comedies, Travelogue, Philosophy, Science, Handbooks, Trials and Sermons (for details see Fanego 1998:89). The samples used for all other subperiods consist of 80,000 words of nonfiction and 120,000 words of fiction.
102 TERESA FANEGO
HC EmodE3 16401710 [117,300 words] LModE1 17001726 [200,000 words] LModE2 17321757 [200,000 words] LModE3 17611797 [200,000 words] LModE4 18501879 [200,000 words]
the Ving NP (hybrid) 19 29 27 9 6
the Ning of-phrase (nominal) 25 27 21 18 35
the N/Ving X (ambiguous) 1 3 8 2 4
Table 1. The-gerunds in British English (16401879)
Total 45 59 56 29 45
following results: c2= 19.70; p 0.001. The distribution is highly significant. The label `ambiguous' is used for structures such as: `the casting up and ruminating on which was my daily and only Pleasure' (COLMOBAENG 1743 Fielding A Journey from This World to the Next, 94). These lack direct objects and, instead, govern dependents that might occur readily in both NP or VP structure (compare `one's reflections on sexism'), hence their categorial status as either NPs or VPs is unclear. In any case, just like the hybrid type, they would be ungrammatical today. Four out of the nine examples of hybrid gerunds recorded in LModE3 (17611797) occur in texts from the early part of the subperiod: Massey's Origin of Letters (1763; 1 ex.) and Sarah Robinson Scott's The History of Sir George Ellison (1766; 3 exx.).
4. An alternative explanation for the decline of the-hybrids A mere glance at Table 1 shows that the changes affecting gerundial structures introduced by the can be dated back to the subperiod I have identified as LModE3 (17611797). Prior to this, the-hybrids (e.g. the gaining her affections) had been slowly but steadily gaining ground,
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 103 and usually served, as already noted, to avoid the use of a bare verbal gerund (e.g. gaining her affections) in the clausal slots of subject, object or predicative. By LModE4 (18501879) hybrids amount to just a small percentage (= some 15%) of all the nominal gerunds recorded in my corpus, their loss being completed in the early years of the twentieth century. Having once clarified that the-hybrids begin to decline in frequency coinciding with the last decades of the eighteenth century, let me just turn briefly to the issue of language standardization. As discussed by Haugen (1966, 1968); Joseph (1987); Pounder (2001); Beal (2004:9093) and others, the process of language standardization follows very similar patterns across European languages. After an initial selection stage in which a specific language variety successfully emerges from the pool of candidate dialects for standardization, there is usually an elaboration stage in which attempts are made to increase the repertoire of formal devices as well as vocabulary appropriate to the new functions of a written standard language. This elaboration stage is followed by what Joseph (1987:108ff.) has labelled the control stage, in which `official or self-appointed controllers (writers, grammarians and other codifiers, teachers etc.) take critical stock of various aspects of the system (vocabulary, syntax, morphology, pronunciation, spelling)' (Pounder 2001:320), such that some elements are praised and others reviled, all in the name of consistency, logic, analogy, clarity, rationality, uniformity and the like. An essential facet of control activity is that variants within the language, often associated with one or more system-motivated changes in progress, `are hierarchized, and sometimes eliminated' (Joseph 1987:109). In the case of English in particular, it is generally agreed (cf. Beal 2004:90) that this control stage coincides in large measure with the decade beginning in 1760, a decade which, as noted by Baugh & Cable (2002:274), `witnessed a striking outburst of interest in English grammar'. In 1761 Joseph Priestley published The Rudiments of English Grammar, which was followed about two months later (February 1762) by Robert Lowth's Short Introduction to English Grammar. The British Grammar, by James Buchanan, appeared in the same year, and a somewhat more elementary manual, by John Ash, was published in 1763 with the title Grammatical Institutes. These were the first in a long tradition of eighteenth- and nineteenth-century English grammars
104 TERESA FANEGO which, in keeping with the defining features of the control stage, as outlined above, claimed that every expression [was] either correct or incorrect and that alternative expressions for the same meaning or function cannot both be correct. In attempting to regulate the vernacular and limit variation in linguistic form, a general inclination prevailed to regard variant forms for the same meaning or function as unacceptable. ... In practice, if not always in theory, grammarians of this period shared a disposition to reject alternative usages as equally correct. If shall is right in this usage, will must be wrong; if among serves several, between must be limited to two (Finegan 1998:54547). Doubtlessly the chief representative of the language controllers that proliferated in eighteenth-century England was Robert Lowth, whose name has become synonymous with prescriptive grammar. Lowth's grammar, published in February 1762, was apparently written with the specific purpose of teaching the correct use of English syntax to would-be as well as established writers (cf. Tieken-Boon van Ostade 2000). It proved immediately popular, so that a second edition of the grammar came out in April 1763. It was followed by forty-six more, twenty-two of which appeared during the eighteenth century (Baugh & Cable 2002:275), and its influence was spread by numerous imitators, including the American expatriate Lindley Murray (Gцrlach 2001:116), who literally took over many passages from it in his well-known English Grammar (1795). Not surprisingly, one of Lowth's prescriptive concerns was gerundial usage. He was the first eighteenth-century grammarian (see Sundby et al. 1991:25557, 36162) to draw attention to some of the constructions that I have discussed earlier in this paper. His comments (1762:11114; emphasis added) deserve quoting in full: The Participle, with an Article before it, and the Preposition of after it, becomes a Substantive, expressing the action itself which the verb signifies: as, `These are the Rules of Grammar, by the observing of which you may avoid mistakes. Or it may be expressed by the Participle, or Gerund; `by observing which:' not, `by observing of which;' nor, `by the observing which:' for either of those two Phrases would be a confounding of two distinct forms. ...
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 105 This Rule arises from the nature and idiom of our Language, and from as plain a principle as any on which it is founded: namely, that a word which has the Article before it, and a Noun, with the Possessive Preposition of, after it, must be a Noun; and if a Noun, it ought to follow the Construction of a Noun, and not have the Regimen of a Verb ... I believe there are hardly any of our Writers, who have not fallen into this inaccuracy. That it is such, will perhaps more clearly appear, if we examine and resolve one or two examples in this kind. `God, who didst teach the hearts of thy faithful people, by the sending to them the light of thy Holy Spirit
:' Collect, Whitsunday. ... Sending is in this place a Noun; for it is accompanied with the Article: nevertheless it is also a Transitive Verb, for it governs the noun light in the Objective Case: but this is inconsistent; let it be either the one or the other, and abide by its proper Construction. ... The Phrase would be proper either way, by keeping to the Construction of the Noun, by the sending of the light; or of the Participle, or Gerund, by sending the light. Lowth's criticism, as can be observed, is aimed precisely at hybrid gerunds such as `by the observing which' or `by the sending to them the light'. This was an area of English grammar where there was clearly divided usage in eighteenth-century England, and, as I pointed out above, a prerequisite to control activity is the existence of some degree of optionality and variation, often associated with a linguistic change in progress in the case under discussion, the massive restructuring of the English complement system as a result of the rise of a new complement type, the verbal gerund, in Late Middle English. As might have been expected given Lowth's prominent intellectual position in eighteenth-century England, his views on the gerund were quickly imitated, when not simply repeated verbatim (e.g. by Lindley Murray 1795:117; cf. Gцrlach 2001:116), in most normative grammars of the period. Visser (1972:§1124) and Sundby et al. (1991) quote statements to the same effect by John Burn (A Practical Grammar, 1766:72), Anselm Bayly (A Plain & Complete Grammar, 1772:72), James Wood (Grammatical Institutions, 1777:103), Lewis Brittain (Rudiments of English Grammar, 1788:124) and many others. Yet with the exception of Visser, whose opinion with regard to the likely influ-
106 TERESA FANEGO ence of normative pressures on the decline of hybrid gerunds has already been quoted in this paper, the prevailing view today is largely the one put forward by van der Wurff (1993, 1997; see also Denison 1998:27172), or more recently by Gцrlach (2001:115) when he asserts that `Lowth's discussion of the verbal noun is eminently clear ? but his guidance does not appear to have had much influence'. That this position is untenable is, however, clear from the data adduced in Table 1: the influence of Lowth and his fellow grammarians was no doubt crucial in bringing about the loss of hybrid gerunds with an initial definite article, and, ultimately, in promoting the diffusion of subjectless verbal gerunds (e.g. gaining her affections) to all clausal functions.11 References Baugh, Albert C., & Thomas Cable. 2002. A history of the English language. 5th edition. London: Routledge. Beal, Joan C. 2004. English in modern times 17001945. London: Arnold. 11 This paper has looked only at British English, but hybrid gerunds, as in (a) below, were also a feature of American English, as could be expected in view of the successive waves of immigrants that moved from Britain to the American colonies during the Colonial period (16071776). The evidence from the COLMOBAENG (see footnote 3 above) suggests that normative pressures also played too a significant role in the decline of the-hybrids in American English, so that by the 1820s they were being consciously avoided in writing, and by the 1850s had completely disappeared (no occurrences are found in a 200,000-word sample of American English from the period 18511879). Several of the prescriptive grammars written in England in which hybrids were stigmatized became popular on the other side of the Atlantic; thus, Lowth's Short Introduction to English Grammar (1762) saw its first American edition in 1775, `and was thereafter shamelessly copied and imitated, serving Harvard students into the 1840s' (Finegan 2001:365). The same applies to Lindley Murray's English Grammar, initially published at York in 1795, but soon having its first American printing in 1800 and passing through some forty editions within half a dozen years. Hybrid gerunds were also condemned by America's pioneer grammarian and lexicographer Noah Webster in A Grammatical Institute of the English Language, Part II (1784:8283), itself the first American grammar to attain wide circulation (Finegan 2001:367). (a) But the making the requisite distinction requires more care of reflection and thought than most men are used to. (COLMOBAENG 1754 Edwards Freedom of the Will,
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 107 Denison, David. 1993. English historical syntax: verbal constructions. London & New York: Longman. Denison, David. 1998. Syntax. In Romaine 1998:92329. De Smet, Hendrik. 2004. Semantics and the variation in complement constructions: gerunds and infinitives following the verb like. Preprints van het Departement Linguistiek 212. Katholieke Universiteit Leuven. De Smet, Hendrik. 2005. For... to-infinitives as verbal complements in Late Modern and Present-day English: between motivation and change. Preprints van het Departement Linguistiek 225. Katholieke Universiteit Leuven. De Smet, Hendrik & Hubert Cuykens. 2005. Pragmatic strengthening and the meaning of complement constructions: the case of like and love with the to-infinitive. Journal of English Linguistics 33, 334. Fanego, Teresa. 1990. Finite complement clauses in Shakespeare's English. Studia Neophilologica 62, 321, 12949. Fanego, Teresa. 1992. Infinitive complements in Shakespeare's English. Santiago de Compostela: Universidad de Santiago de Compostela. Fanego, Teresa. 1996a. The development of gerunds as objects of subject-control verbs in English (14001760). Diachronica 13, 2962. Fanego, Teresa. 1996b. The gerund in Early Modern English: evidence from the Helsinki Corpus. Folia Linguistica Historica 17, 97152. Fanego, Teresa. 1996c. On the historical development of English retrospective verbs. Neuphilologische Mitteilungen 97, 7179. Fanego, Teresa. 1997. On patterns of complementation with verbs of effort. English Studies 78, 6067. Fanego, Teresa. 1998. Developments in argument linking in Early Modern English gerund phrases. English Language and Linguistics 2, 87119. Fanego, Teresa. 2004a. On reanalysis and actualization in syntactic change: the rise and development of English verbal gerunds. Diachronica 21, 555. Fanego, Teresa. 2004b. Some strategies for coding sentential subjects in English: from exaptation to grammaticalization. Studies in Language 28, 32161. Fanego, Teresa. 2006. Drift and the development of sentential complements in British and American English from 1700 to the present day. In Javier Pйrez-Guerra et al. (eds.), Selected papers from the
108 TERESA FANEGO Second International Conference
on the English Language in the Late Modern Period (LMEC2), University of Vigo, 2527 November 2004. Bern: Peter Lang. Finegan, Edward. 1998. English grammar and usage. In Romaine 1998: 53688. Finegan, Edward. 2001. Usage. In John Algeo (ed.), The Cambridge history of the English language. Vol. 5, 358421. Cambridge: Cambridge University Press. Fischer, Olga. 1988. The rise of the for NP to VP construction: an explanation. In Graham Nixon & John Honey (eds.), An historic tongue: studies in English linguistics in memory of Barbara Strang, 6788. London & New York: Routledge. Fischer, Olga. 1989. The origin and spread of the accusative and infinitive construction in English. Folia Linguistica Historica 8, 143217. Fischer, Olga. 1995. The distinction between to and bare infinitival complements in late Middle English. Diachronica 12, 130. Gцrlach, Manfred. 2001. Eighteenth-century English. Heidelberg: Carl Winter. Haugen, Einar. 1966. Language conflict and language plan
ning: the case of modern Norwegian. Cambridge, MA: Harvard University
Press. Haugen, Einar. 1968. The Scandinavian languages
as cultural artefacts. In Joshua A. Fishman, Charles A. Ferguson & Jyotitindra Das Gupta (eds.), Language problems of developing nations, 26784. New York: Wiley. Hoekstra, Teun. 1986. Deverbalization and inheritance. Linguistics 24, 54984. Joseph, John Earl. 1987. Eloquence and power: the rise of language standards and standard languages. London: Frances Pinter. Kytц, Merja. 1996. Manual to the diachronic part of the Helsinki Corpus of English Texts: coding conventions and lists of source text
s. 3rd edition. Helsinki: Department of English, University of Helsinki. Los, Bettelou. 1998. The rise of the to-infinitive as verb complement. English Language and Linguistics 2, 136. Los, Bettelou. 2005. The rise of the to-infinitive. Oxford: Oxford University Press.
THE LOSS OF HYBRID GERUNDS IN MODERN ENGLISH 109 Lowth, Robert. 1762. A short introduction to English gramma. London: J. Hughs for A. Millar and for R. & J. Dodsley. (Facsimile reprint from Lowth's own copy, with a notice by Robin C. Alston. Menston: Scolar Press, 1968.) Mair, Christian. 2002. Three changing patterns of verb complementation in Late Modern English: a real-time study based on matching text corpora. English Language and Linguistics 6, 10531. Mair, Christian. Forthcoming. The 19th century as a critical period in the formation of the Modern English system of non-finite complement clauses: the case of remember. In Merja Kytц, Mats Rydйn & Erik Smitterberg (eds.), Nineteenth-century English: stability and change. Cambridge: Cambridge University Press. Miller, D. Gary. 2002. Nonfinite structures in theory and change. Oxford: Oxford University Press. Moessner, Lilo. 1997. The history of English -ing-constructions. History of English (Seoul) 3, 10520. special issue
for Professor Kim In-Sook. Murray, Lindley. 1795. English grammar, adapted to the different classes of learners. York: Wilson, Spence & Mawman. (Facsimile reprint. Menston: Scolar Press, 1968.) Pounder, Amanda. 2001. Adverb-marking in German and English: system and standardization. Diachronica 18, 30158. Rohdenburg, Gьnter. 1995. On the replacement of finite complement clauses by infinitives in English. English Studies 76, 36788. Rohdenburg, Gьnter. 1996. Cognitive complexity and increased grammatical explicitness in English. Cognitive Linguistics 7, 14982. Romaine, Suzanne (ed.). 1998. The Cambridge history of the English language. Vol. 4. Cambridge: Cambridge University Press. Rudanko, Juhani. 1998. Change and continuity in the English language: studies on complementation over the past three hundred years. Lanham NY & Oxford: University Press of America. Rudanko, Juhani. 1999. Diachronic studies of English complementation patterns: eighteenth century evidence in tracing the development of verbs and adjectives selecting prepositions and complement clauses. Lanham NY & Oxford: University Press of America. Rudanko, Juhani. 2000. Corpora and complementation: tracing sentential complementation patterns of nouns, adjectives and verbs over the last three centuries. Lanham NY & Oxford: University
110 TERESA FANEGO Press of America. Sundby, Bertil, Anne Kari Bjшrge & Kari E. Haugland. 1991. A diction- ary of English normative grammar 17001800. Amsterdam & Philadelphia: John Benjamins. Tajima, Matsuji. 1985. The syntactic development of the gerund in Middle English. Tokyo: Nan'un-do. Tieken-Boon van Ostade, Ingrid. 2000. Robert Dodsley and the genesis of Lowth's Short introduction to English grammar. Historiographia Linguistica 27, 2136. Visser, F. Th. 1972. An historical syntax of the English Language. Vol. 2. Leiden: E. J. Brill. Vosberg, Uwe. 2003. The role of extractions and horror aequi in the evolution of ing complements in Modern English. In Gьnter Rohdenburg & Britta Mondorf (eds.), Determinants of grammatical variation in English, 30527. Berlin & New York: Mouton de Gruyter
. Vosberg, Uwe. 2004. Determinanten grammatischer Variation: Verschiebungsprozesse bei satzwertigen Komplementstrukturen im Neuenglischen. Dissertation, University of Paderborn. Wagner, Susanne. 2002. Non-finite verb complementation in the 17th, 18th and 19th centuries. A corpus-based study. Lizenziats-Arbeit der Philosophischen Fakultдt der Universitдt Zьrich. Warner, Anthony. 1982. Complementation in Middle English and the methodology of historical syntax. London & Canberra: Croom Helm. Webster, Noah. 1784. A grammatical institute of the English language. Part 2. Hartford: Hudson and Goodwin. (Facsimile reprint. Menston: The Scolar Press, 1968.) Wurff, Wim van der. 1993. Gerunds and their objects in the Modern English period. In Jaap van Marle (ed.), Historical linguistics 1991, 36375. Amsterdam & Philadelphia: John Benjamins. Wurff, Wim van der. 1997. Gerunds in the Modern English period: structure and change. History of English (Seoul) 3, 16396. Special Issue for Professor Kim In-Sook. Yoon, James Hye Suk. 1996. Nominal gerund phrases in English as phrasal zero derivations. Linguistics 34, 32956.