Date: Tue, 27 Aug 2002 11:01:59 -0700
From: David Golumbia
Subject: Helasvuo (2001) Syntax in the Making: The Emergence of Syntactic Units in Finnish Conversation
Helasvuo, Marja-Liisa (2001) Syntax in the Making: The Emergence of Syntactic Units in Finnish Conversation. John Benjamins Publishing Company, xiv+175pp, hardback ISBN 1 55619 394 7 (US), $73.00, 90 272 2619 9 (Eur), Studies in Discourse and Grammar 9.
Announced at http://www.linguistlist.org/issues/13/13-325.html
David Golumbia, Long Beach, CA
In a series of carefully linked empirical demonstrations and theoretical arguments, Helasvuo (H) develops a model for syntactic structure that touches neatly on themes raised by Paul Hopper, Sandra Thompson (especially Hopper 1987, 1988; Hopper and Thompson 1980, 1984), Barbara Fox (1987), M. A. K. Halliday, Talmy Givón, and other leading functional theorists. Combining her own original research (on Finnish conversational corpora) with thoughtful examinations of main threads in recent theoretical literature, H develops a relatively complete rethinking of the basis of grammatical phenomena that is nevertheless solidly grounded in several significant research programs.
1. Introduction (1-18). H states that she takes the clause as the basic unit of study but not necessarily of syntax, following what appears to her the emergence of regular structures in Finnish discourse. In this she follows Hopper's (1987, 1988) well-known theory of Emergent Grammar with its emphasis on "recurrent partials" (Hopper 1988, 118; H 1). Taking a broader view of Hopper's meaning than have some other researchers, H declares her "aim is to study the grammaticized 'recurrent partials' in their natural environments of use" (1). In this view grammar is a "vaguely defined set" of "grammaticized recurrent patterns of discourse" (2). H rightly notes her innovative approach in being concerned not "with the history of the grammaticized items or constructions, per se, but rather, attempts to describe their grammar as it is being realized in present-day Finnish discourse" (2).
Following Langacker (1987), H presents her conversational data in terms of schemas, "constantly evolving set[s] of cognitive routines that are shaped, maintained,, and modified by language use" (Langacker 1987, 57; H 4). H's data are fascinating and suggestive because they include a great deal of overlapping interaction, fragments, stop-starts, and so on: in other words, they seem very close to everyday conversation.
2. The Dynamics of the Clause (19-84). H explains her focus on spoken corpora, but she is careful to relate this focus to phenomena found in written language as well: in both cases she is convinced that traditional descriptions of grammatical rules do not fit the data precisely enough. H searches for a principle we might call weakly functionalist, in that she searches for certain kinds of semantic or pragmatic conditioning to account for some broad grammatical patterns. Focusing on the clause, and within it the role of word order, she observes that the category of person is the most important one in determining overall discourse structure and to a great extent grammatical structure. She suggests that "in previous research, the importance of personal pronouns has been largely underestimated" (82) considering their prevalence in discourse; "personal pronouns should be taken as a central force in the structuring of argument relations in discourse."
In particular, 1st and 2nd person pronouns provide core cases for clause structure, providing schemas that "do important discourse work in the marking of Noun Phrases (NPs) that serve to track participants in discourse" (83). Thus the interplay of 1st and 2nd person pronouns create agreement patterns that influence 3rd persons and even oblique cases. H is careful to point out that this conclusion applies to her Finnish data in particular, and that "languages respond in different ways to the various discourse needs that speakers are faced with," citing Mandarin and Nuu-chah-nulth as examples of languages in which "the clause does not emerge so clearly as a level of syntactic organization."
3. Grammaticization of the Subject Role (85-104). Following her observations on the clause, H continues to explore the primacy of the Subject role and Subject/Object relations, again emphasizing the use of personal pronouns to provide what looks like grammatical structure. First, she explores whether there is in fact a "unified subject role" (85) in Finnish from a statistical perspective (in terms of the occurrence of 1st and 2nd person pronouns, other pronouns, NPs, and so on. She determines that "there is an overwhelming tendency in spoken discourse for NPs in the transitive subject role" (87, following terminology of Du Bois 1987 and elsewhere). Further analysis of the data reveals that there is in fact a narrow definition of syntactic Subject that seems unified with discourse Subject, which "thus defined almost always takes a preverbal position" in addition to being highly correlated with personal pronoun occurrence.
H's analysis extends from the core cases of simple transitive and intransitive clauses to the complex oblique cases and ergative agreement also found in Finnish (which is part of what makes Finnish conversational data an especially interesting target for study). This allows her to extend her analysis from the prevalent uses of structuring parts of speech to more peripheral constructions, and therefore to the kinds of animacy and person hierarchies that are becoming familiar in a range of linguistic approaches, in general directly emerging from or in response to Silverstein (1976). H notes that "new referents are rarely introduced in the subject role" and that "there is a difference between new mentions made in the core roles and new mentions in the oblique role: those referents that are introduced in the oblique role are hardly ever mentioned again in subsequent discourse, whereas referents that are first mentioned in the core roles are much more likely to be mentioned again, i.e. to be tracked in the discourse" (93). Finnish includes an existential clause schema (E-NP) and partitive "Subject" construction that "lack most of the coding features that characterize subjects: they do not trigger agreement, and E-NPs may appear in the partitive case as well as the nominative. . The referents of NPs are usually human. Thus, their respective discourse profiles are completely different, and so are the coding features. We can conclude that they do not represent the same syntactic function, i.e. the subject" (103).
4. Free NPs (105-132) Moving from instances that feature the most prominent available grammar (transitive and intransitive NPs and clauses), H now turns to what appear to be among the least structured elements, namely Free NPs, again in the same corpora she has been examining throughout. "In Finnish there is agreement not only between the subject and the predicate, but also, within the noun phrase between the head and its modifiers. This is what Lehmann (1988) calls 'internal agreement.' . This is one of the defining features of noun phrase formation in Finnish" (105). H argues that free NPs serve a limited set of discourse and syntactic functions in Finnish and help us to gain a clearer picture of overall grammatical operations in the language, since they occur frequently in the spoken corpora.
H uses a relatively strict method to determine what counts as a free NP in her data (106-8). Addressing the issue in the functionalist tradition directly, H argues that free NPs are "a type of syntactic unit which is distinct from and not reducible to clauses" (113). The elaborate case marking system of Finnish is applied selectively to free NPs, and they serve a variety of discourse functions. Following suggestions of Hopper and Thompson (1980, 1984), H suggests that "referential free NPs can be used to help to highlight a referent or to focus on a referent that is a member of a larger set of referents already under discussion" (131). "Among the predicating free NPs, identifying and classifying free NPs function either to characterize referents or to disambiguate the intended referent(s). IN other words, they serve mainly in negotiating reference, whereas constructions with a theme and orientation make predications that may initiate something new, rather than look back to check the understanding of prior talk." Again, H's work demonstrates the power of something like participant role to structure significant amounts of linguistic production.
5. Intonation and Syntactic Structuring (133-150). In this final chapter H's work again takes an innovative turn; she has not previously focused on phonetic or prosodic issues, despite the fact that the data she has been offering (as is the case in many works) seems to invite such analysis. H turns to this issue directly, focusing on intonation and its interaction with the emerging principles she has identified as syntactic structuring. Again following provocative views of Langacker (1997), H looks for ways in which intonation units and apparent syntactic constituents do and do not overlap. This allows for a flexible, many-layered approach to the idea of "constituent" itself that crosses phonological and morphological levels: "NPs are rarely split across intonation units. . Most of the time intonational phrasing is convergent with syntactic phrase structure" (140).
Turning to VPs and oblique NPs, H finds the Finnish evidence more ambiguous. "Oblique NPs show a certain vagueness in regard to their syntactic integration. It is no wonder that they often form intonation units of their own" (145). In the core cases, on the other hand, strong correlations are the rule. "In terms of intonational groupings, the bond between the subject and the verb seems to be even stronger than that between the verb and the object, b4ecause objects are more often in a different intonation unit from the verb than are subjects" (148). H agrees that in Finnish, as in many other languages, the category of VP "may be problematic." The tight reinforcement of grammar and intonation in the core cases suggests that "syntactic relations appear most clearly in the clause core" (149) which further "shows the robustness of the category NP: as we saw in the analysis, NPs that are clausal constituents are rarely split into two intonation units. Free NPs form syntactic units of their own, and they are most often also produced in intonation units of their own."
6. Conclusion (151-154). H reflects on her investigation by noting again that "it is the clause core where grammatical relations appear most clearly" (151). She notes that "case marking, agreement, and word order . highlight the role of the clause core as the locus for the most explicit coding." Summarizing the work of the latter chapters in particular, she writes that "if an argument was produced in a different intonation unit from the predicate, it was more likely to be the object than the subject. This patterning suggests a subject-verb grouping on a par with the traditional verb-object grouping. Thus, there is evidence in the data for a more flexible analysis of grammatical constituency than has traditionally been assumed, allowing for different kinds of groupings among the elements" (153). Furthermore, though "there is strong evidence for the clause as an emergent grammatical unit in Finnish," H would "by no means like to suggest that this would be true of languages in general. In fact, we have discourse studies from other languages that point to the fact that this is not the case."
This is an extremely strong, carefully thought-out, insightful and reflective book. It clearly reflects many years of thought, condensed into an easy-to-follow but dense text. It contains much to interest linguists of whatever stripe, especially those with interests in the foundations of syntactic theory, cross-linguistic variation in basic structuring patterns, patterns of transitivity and ergativity, functional theories, discourse theories, corpus linguistics - in short, many subjects for a book that appears relatively modest at first glance.
Nevertheless, the topics feel neither glossed over nor ornamental; their relation to H's core argument is clear both from her text and, on reflection, in the literature and the data themselves. In other words the phenomena she is describing, and the arguments she teases out of the theoretical literature, seem to emerge organically, and one often finds her addressing just the issues that need to be touched on.
This makes her approach both synthetic and analytic, and it helps that her focus is on a small body of corpora from a single language that she clearly knows well, so that she can pay special attention to the apparently para-linguistic phenomena (stop-starts, breaks, turns, repairs, and so on) that do seem critical to any linguistic function and are too often under-analyzed. That she is able to read through such data to a coherent and meaningful thesis is achievement enough, but that it reflects so suggestively on so many strands of other recent work makes this an especially valuable book.
The book displays a mastery of the functional literature in particular, and is among the best existing studies of a single set of language phenomena from what H rightly calls an emergent perspective (in the narrow sense of Hopper 1987, 1988): allowing the data themselves to speak clearly enough to shape formal and theoretical observations. Her focus on the "core cases" in her discourse samples feels right: what is it that people have routinized, do automatically, or rely on heavily and/or paradigmatically when they speak?
Because of this, the depth of her functional analysis do not really bar her analysis from relevance to other theoretical approaches. Indeed, much of her concentration on the core cases seems in line with generative grammar, since something like these cases seem homologous to the kinds of functional projections and core operations posited in recent generative work. If there is one shortcoming in the book, then, it would be the volume's brevity: despite her proper focus on the core cases, one suspects H has a lot to say about the peripheral, "broken" and marginal parts of language production and their relation to grammar, and it is provocative to imagine her analysis extended to these areas.
Du Bois, J. (1987). The Discourse Basis of Ergativity. Language 63, 805-855.
Fox, B. (1987). Discourse Structure and Anaphora: Written and Conversational English. New York: Cambridge University Press.
Hopper, P. (1987). Emergent Grammar. In Berkeley Linguistics Society. Proceedings of the 13th Annual Meeting. Berkeley, CA: Berkeley Linguistics Society, 139-157.
Hopper, P. (1988). Emergent Grammar and the A Priori Grammar Postulate. In Tannen, D. (ed.) Linguistics in Context: Connecting Observation and Understanding. Advances in Discourse Processes, Volume 29. Norwood, NJ: Ablex Publishing, 117-134.
Hopper, P. and Thompson, S. (1980). Transitivity in Grammar and in Discourse. Language 56, 251-299.
Hopper, P. and Thompson, S. (1984). The Discourse Basis for Lexical Categories in Universal Grammar. Language 60, 703-752.
Langacker, R. (1987). Foundations of Cognitive Grammar, Vol. I. Stanford: Stanford University Press.
Langacker, R. (1997). Constituency, Dependency, and Conceptual Grouping. Cognitive Linguistics 8, 1-32.
Lehmann, C. (1988). On the Function of Agreement. In Barlow, M. and Ferguson, C. A. (eds.) Agreement in Natural Language: Approaches, Theories, Descriptions. Stanford: CSLI. 55-65.
Schegloff, E. (1991). Conversation Analysis and Socially Shared Cognition. In L. Resnick, L., Levine, J., and Teasley, S. (eds.) Perspectives on Socially Shared Cognition. Washington, DC: American Psychological Association. 150-171.
Silverstein, M. (1976). Hierarchy of Features and Ergativity. In Dixon, R. M. W. (ed.) Grammatical Categories in Australian Languages. Canberra: Humanities Press. 112-171.