LINGUIST List 12.1341

Wed May 16 2001

Review: van Gelderen, Hist of Eng Reflexive Pronouns

Editor for this issue: Terence Langendoen <terrylinguistlist.org>


What follows is another discussion note contributed to our Book Discussion Forum. We expect these discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in.

If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for discussion." (This means that the publisher has sent us a review copy.) Then contact Simin Karimi at siminlinguistlist.org or Terry Langendoen at terrylinguistlist.org.


Directory

  • Claudia.Lange, van Gelderen, History of English Reflexive Pronouns

    Message 1: van Gelderen, History of English Reflexive Pronouns

    Date: Wed, 16 May 2001 21:05:13 +0100
    From: Claudia.Lange <Claudia.Langemailbox.tu-dresden.de>
    Subject: van Gelderen, History of English Reflexive Pronouns


    van Gelderen, Elly (2000) A History of English Reflexive Pronouns: Person, Self, and Interpretability. John Benjamins, hardback, xiv, 277 pp., Linguistics Today 39

    Claudia Lange, Technische Universitaet Dresden

    Synopsis: Van Gelderen has repeatedly dealt with reflexive and emphatic SELF in the history of English from a generative perspective. In her last monograph on the subject, adopting the Minimalist framework, she attempts nothing less than to link "the changes in reflexives to the transformation of English from a synthetic to an analytic language" (p. 1), or, to be more specific: "... I argue that languages and different stages of the same language differ as to which features are Interpretable. [...] the status of features ultimately accounts for differences in word order, Case and agreement across languages, and for whether a language is synthetic or analytic. It will also account for the referential or non-referential nature of pronouns" (p. 12). While it is in line with current generativist thinking to locate parametric differences between individual languages as well as differences between historical stages of the same language to the rise of new Functional Categories (FCs) and the changing status of features, tying the emergence of a clear division between pronouns and anaphors to the same process is a novel approach.

    The book consists of 7 chapters, preceded by an introduction in which van Gelderen specifies the theoretical framework underpinning the book, namely Minimalism (Chomsky 1995) with its theory of feature checking. Briefly and simply, each lexical item carries grammatical features which have to be checked in the course of a derivation, with strong features triggering overt movement. Features are further either Interpretable, that is relevant to the interpretation at the level of Logical Form (LF), or Non-Interpretable at LF; they exist to trigger movement to a higher FC to become checked and erased. In Modern English, case and all verbal features other than tense (person and number) are Uninterpretable, while in Old English (OE), case features are Interpretable. As the quote above already indicates, van Gelderen crucially makes use of the notions of feature strength and feature Interpretability to motivate her analysis. Reinhart and Reuland's (1993) Chain Condition is another theoretical prerequisite which figures prominently in van Gelderen's account. It states that "A maximal A- chain (a1...an) contains exactly one link -- a1 -- that is both +R and case marked" (p. 696) and licenses the use of plain pronouns as reflexive anaphors in case they are -R, that is not referential. The property of being referential is defined as involving "a full specification for phi-features and structural Case" (p. 697), so assuming that OE pronouns have Inherent case makes those pronouns referentially defective, which is why they can function as reflexives. The remainder of the book is devoted to tracing the loss of Inherent Case and the changes in feature composition throughout the history of English and to linking these processes to the rise of new reflexive markers.

    The first two chapters provide a detailed text- based study of reflexivity in Old and Middle English. Chapter 1 gives a descriptive account of the way reflexivity was expressed in OE. It tackles two issues: first, the reflexive use of the personal pronoun, which contradicts the Binding Principles and is therefore in need of explanation within the framework adopted, and second, the use of the 'emphatic' adjective SELF. Van Gelderen examines different texts in turn, most extensively Beowulf. More data to illustrate reflexivity in OE come from the Junius manuscript, the Exeter Book and for late OE from Alfred's Pastoral Care and AElfric's Homilies. The picture that emerges from examination of these texts is the following: the accusative personal pronoun regularly serves as the reflexive pronoun. SELF appears predominantly in attributive position modifying mainly nominal subjects rather than pronouns; it is typically used 'emphatically'. In the later OE texts, SELF continues to be used as modifier providing emphasis, but also increasingly postmodifies personal pronouns used reflexively. The question of whether dialectal differences are responsible for the change in distribution of SELF is then addressed by comparing two interlinear Bible glosses: the Northumbrian (Northern) Lindisfarne Gospels and the Mercian Rushworth Glosses. Van Gelderen concludes that the more northern and earlier texts used SELF more sparingly, but if so, SELF tended to be used more frequently with a reflexive. Thus, the situation in OE is consistent with R&R's Chain Condition. The picture begins to change, however, when the development in Middle English (ME) is taken into account.

    Chapter 2 proceeds from early ME with special emphasis on Layamon's Brut, other texts taken into account are The History of the Holy Rood Tree and Hali Meidhad. For later stages of the language, van Gelderen discusses the Gawain Poet and Chaucer. She deals first with the grammaticalization of SELF during this period, exemplifying the relevant changes with examples from the two manuscript versions of Layamon. It turns out that the new compound reflexive pronoun+SELF appears first in the third person while first and second person continue to be used reflexively. Her explanation for this person split makes reference to differences in feature specification: "In older English, first and second person pronouns have fewer fully marked phi-features (e.g. for person and number) and can therefore continue to function anaphorically even though they lose Inherent Case; third person pronouns, on the other hand, have fully marked phi-features and when Inherent case disappears, they cease to function anaphorically"(p.97).

    In chapter 3,"Pro-drop and feature strength" van Gelderen presents evidence that OE had pro-drop of at least four different kinds: with infinitives ("big PRO"), with expletive or non-referential subjects, and topic-drop. Her emphasis is on showing that, unlike generally assumed, referential pro- drop exists in OE. Her overall aim is to use the occurrence of pro-drop as an indicator of why third person pronouns continue to be used reflexively even after the loss of Inherent Case: since, as she shows, pro-drop is more common with third person subject pronouns, this is taken to prove that there is a difference in feature specification: third person pronouns are more specified in virtue of being more deictic and can therefore license pro-drop. Pro-drop continues to occur in English, as is demonstrated in the remainder of the chapter. She concludes that subject pro-drop is not productive after the seventeenth century, which for her indicates that by that time, person and number features are unspecified, coinciding with the demise of the simple pronoun as reflexive marker.

    Chapter 4, "The loss of verbal agreement and verb-movement", extends her argument even further by linking the occurrence of pro-drop in different texts to the degree these texts display progressive loss of verbal agreement. Agreement in the third person is kept up longer than for first and second person, again taken as an indicator of different feature strength. Additionally, van Gelderen argues that "with the loss of inflections, features become Uninterpretable and FCs are introduced"(p. 152).

    Chapter 5 is devoted to "The Loss of Inherent Case". In chapter 1, van Gelderen explained the use of simple pronouns as reflexives with reference to the notion of Inherent Case: since inherently Case marked elements are not referential (-R) for purposes of binding, they satisfy Reinhart & Reuland's Chain Condition and can be used reflexively. In this chapter, she provides evidence for the existence of Inherent Case in OE; points in favour of that assumption are: rich morphological case marking, case assignment according to the theta-role of the verb and/or preposition, retention of cases under movement (passivization). According to her, morphological case continues to be assigned to third persons up to early ME even after it ceases to be assigned to first and second person by late OE.

    In chapter 6, "Ergativity and the person split", van Gelderen examines constructions with non-nominative Agents or Experiencers such as the impersonal, which she labels 'ergative'. With the focus once more mainly on 'Beowulf', she finds a person split again in that third person pronouns in possessive constructions with 'be' take dative case while first and second person pronouns are assigned the nominative. She concludes that "it can be argued that third person pronouns continue to be assigned thematic (inherent) Case by the verb and need not move to a Specifier of an FC to check Case (p. 239) ... Therefore in Old and early Middle English, the demise of impersonals and the loss of inherent Case are related because in both cases, first and second person pronouns are the first to experience the loss."

    Summing up, the seventeenth century saw the end of pro-drop as well as the demise of the simple pronoun used reflexively, both developments being due to the change to Structural Case and to person and number features having become unspecified by that time (p. 136). The change from a synthetic to an analytic language goes along with an increase in Non- Interpretable features. A synthetic language is more likely to use plain pronouns as reflexives because their features are still strong, that is Interpretable. When the language becomes more analytic, with "a general increase in Uninterpretable features"(p. 247) personal pronouns can no longer be used reflexively.

    Critical evaluation: The book provides a wealth of data which is extremely useful for everybody interested in the topic regardless of theoretical orientation. The discussion is based on a meticulous examination of texts and illustrated with an abundance of examples. The problem of which texts to choose is a notorious one in all historical linguistics and particularly prevalent when one uses texts counts and statistics in order to prove a point, as van Gelderen does. I would just like to note that for both OE and ME, secular sources such as chronicles, wills and laws are not taken into account, which could have changed the picture since they are for the most part not translated from Latin. It is further dubious whether her table on p. 66, taken to illustrate the decline of emphatic SELF, has any empirical value as the texts compared are from wildly different genres. Her chronology of texts is also on shaky grounds; Alfred's Pastoral Care should certainly not be grouped under late OE, and although van Gelderen acknowledges that the two Layamon manuscripts are now thought to be from more or less the same time, she still bases most of her argument for the grammaticalization of pronoun + SELF on precisely the development from one manuscript to the other. Van Gelderen claims that around 1250, SELF underwent grammaticalization and became a noun first in combination with a third person pronoun; she assumes that the internal structure of pronoun + SELF changed from a noun modified by an adjective to a determiner with SELF as nominal head. Her claim that SELF became a noun around that time is based on exactly one example (p. 104) from the OED, dated to 1300. The next instance listed in the OED is from 1472, casting doubt on the assumption that nominal SELF was already well established at that time. Relying on the notion of grammaticalization to account for the unexpected emergence of the compound reflexive raises a theory-internal problem: grammaticalization theory fundamentally makes use of the notions of gradualness and functionalism in order to account for linguistic change, notions which so far had no place in generative diachronic syntax (but see van Kemenade (1999) for an attempt at reconciliation).

    Another theory-internal problem concerns the relation of (abstract) Case to (surface) case: to assume that OE had Inherent Case is more or less uncontroversial, so chapter 5 of the book is not telling a new story; the question which so far has not satisfactorily been answered is how morphological (surface) case and (abstract) Case are related (cf. van Kemenade/Vincent (1997)). Van Gelderen, in tabulating the progressive loss of inflectional morphology, seems to take a one-to-one relationship between case and Case for granted and does not discuss this fundamental issue any further. Another point where the great merit of the study, the detailed investigation of actual texts, clashes with the theoretical framework is the purported variability of feature strength: "In summary, Old English has pro-drop, especially with third person and slightly more with singular than with plural, which is not unexpected if the third person features are specified. One of the problems is that the presence or absence of features is never absolute: the strength of the features can vary" (p. 137). This seems to me a big problem indeed for a theory which standardly stipulates binary divisions and strict categories. Further, allowing variable feature strength makes the whole study meaningless in the sense of unfalsifiable unless some principled account of the parameters of variation is provided. A more general point is that to my knowledge, there is no crosslinguistic evidence that the synthetic or analytic character of a language and the form of the reflexive are in any way related, which casts some doubts upon the analysis as a whole.

    References Chomsky, Noam (1995) The Minimalist Program. Cambridge, Mass.: MIT Press.

    K´┐Żnig, Ekkehard & Peter Siemund (1999) "Intensifiers and Reflexives: A Typological perspective." In: Frajzyngier, Zygmunt & Traci S. Curl (eds.), Reflexives. Forms and Functions. Amsterdam: John Benjamins. 41-74.

    - -- (2000) "The Development of Complex Reflexives and Intensifiers in English." Diachronica XVII: 1.39- 84.

    Reinhart, Tanya & Eric Reuland (1993) "Reflexivity." Linguistic Inquiry 24.4: 657-720.

    van Gelderen, Elly (1999) "Bound Pronouns and Local Anaphors.". In: Frajzyngier, Zygmunt & Traci S. Curl (eds.), Reflexives. Forms and Functions. Amsterdam: John Benjamins. 187-225.

    van Kemenade, Ans & Nigel Vincent, eds. (1997) Parameters of morphosyntactic change. Cambridge: Cambridge University Press.

    van Kemenade, Ans (1999) "Functional categories, morphosyntactic change, grammaticalization." Linguistics 37-6: 997-1010.

    Claudia Lange is junior lecturer in Linguistics at the Technische Universitaet Dresden. She is working on her PhD on SELF in the history of English.