Abstract
This paper argues that the ‘nonfinite’ contexts in which negation is preverbal in Bangla are not directly correlated with the possibility of tense inflection, agreement, or the availability of subject licensing positions. Rather, we must make the special word order depend on the difference between deictic anchoring and anaphoric dependence of the clause’s temporal information. This specific view of finiteness is therefore different from the kind of finiteness that the licensing of overt subjects is sensitive to, but is related rather to the notion of independent assertability. In addition, this commentary argues that we need to move towards a Direct Linearization view of deriving word order effects. The solution proposed here is argued to provide a simpler and more elegant statement of the word order pattern, without using word order movements such as head-movement, roll-up and remnant movements that have no feature checking motivations.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
In the Bangla examples in the text, the digraph ‘C(onsonant)h’ is used to express the aspirated version of the consonant C, ṭ, ḍ and ṛ are retroflex versions of the dental series t, d and r. The abbreviations used are as follows: Past, Pres, Fut for past present and future tense respectively; 1, 2, 3 indicate verbal agreement for the person of the subject; cop for the copula; neg—negation; class—classifier; perf perfective, cpart the conditional participle; ppart the perfect participle; ipart the imperfective participle; ger the gerund; nom, acc, dat, gen for nominative, accusative, dative and genitive case respectively. The nominative is unmarked, except in pronouns, and case is only annotated in the examples where explicit morphology indicates it.
The imperfective participle occurs in a number of different distributional contexts: in complement position, as a purpose clause and as a temporal adjunct. Only in the latter construction is an overt lexical subject possible. Once again, see Biswas (2013) for discussion. Concerning the perfective participle, Van der Wurff (1989) discusses the Bangladeshi dialects of Bangla in which a conjunctive participle may actually have an overt nominative subject unlike in the Indian varieties. But even in these dialects the phenomenon is restricted, and in particular does not seem to extend to volitional nominative subjects (Dasgupta pc).
Dasgupta (1996) claims that subjunctive forms simply do not allow post-verbal elements at all, contrasting with their indicative counterparts. However, Dasgupta (p.c.) informs me that this overstates the case, and that examples can indeed be found if the context is set up correctly. True uninflected participles/infinitives do seem to exclude the post-verbal position, however. I will leave the analysis of these word order facts for future research.
See also Giorgi (2010) for an analysis of the subjunctive tenses in Italian as ‘agreeing’ tenses.
In many languages of course, subjunctive morphology is distinct from indicative morphology, showing that it is perfectly possible for languages to mark the distinction between independent anchoring and anaphoric anchoring morphologically.
The ‘negative polarity’ items used here are Wh in-situ elements in the language—K-words. They are used in situ to make Wh questions; with an emphatic particle added they can be interpreted as free choice items in certain non-negated verb forms; when c-commanded by negation they get an obligatory narrow scope indefinite interpretation like NPI any in English.
There are a number of Neo-Brodyan proposals in the literature at the moment (Adger et al. 2009; Adger 2010; Bye and Svenonius 2011) which share some features with what I will be assuming here, and which I consider to belong to the family of DLTs. The ‘best’ DLT, i.e. the one that can elegantly account for language variation in word order while building in the asymmetries we find, is still a matter of ongoing investigation. The proposal I make in this short paper is an attempt to state a maximally elegant DLT for this particular set of Bangla facts, and is therefore a contribution to the more general programme.
Placing the @ sign at the top of the tree would give you SVO, and placing it even higher would give you VSO languages, retaining the relative order of arguments.
To be perfectly DLT, however, we might even want to state this as a direct linearization statement rather than go through the intermediary of a syntactic placeholder, for example:
Bangla TP Linearization Statement: “Within a morphological word zone, linearize specifiers top down first, and the morphological word last.” I leave such radical implementations, as well as the impact of chunking by means of phases, to further investigation.
It is not my intention here to eliminate movement from the syntactic system. Criterial feature checking remains, and only purely word order movements are eliminated. The linearization algorithm is intended to follow all such operations of the syntactic computation, since it essentially translates the symbolic representation across the interface into the auditory domain and is therefore an interface mapping rule. I also abstract away here from the issue of phases. It may also be the case that linearization proceeds in phases, and that would also imposes order over and above the statements made here. Phases may be universally set, or language-specific. There is no space in this short paper to explore the various options.
This use of the asterisk differs from the Bye and Svenonius use, in that the latter use it to indicate that all the heads below it (until the next asterisk is met) form a Brodyan word with the asterisked head. Moreover, Bye and Svenonius do not discuss the possibility of ‘lowering’ and therefore the asterisk functions as an indication of the spell-out position of the word as well, making it even more similar to head movement.
See Ramchand (2005) for a discussion of the distribution and meaning of ni when compared to na as sentential negation.
The other way to do it is to assume that Fin is not present in nonfinites at all, but then we would have to make the * diacritic on Neg sensitive to whether the immediately subordinate head is Fin, or whether it is T or something lower—only in the former case do we want word formation to kick in. Alternatively, we could adjust the notation so that it is the lower head which determines whether the higher one combines with it via word formation or not. I stick to the conservative implementation above since there is no scope in this article for deciding between different technologies with this limited data set. I put the issue aside for further research, noting only that the decisions on how to implement the generalization expressed here will hinge on architectural decisions about the universality of the functional sequence and the omissability of heads.
Simpson and Syed (2013) point out that there are other elements in Bangla that also show this variable positioning between a sentence-final position and a roughly second position location. These include the question particle ki and the subjunctive subordinator jodi- ‘if’. They mention these facts to lend plausibility to their analysis that lexical items in Bangla can spell out either in head (final) or (leftward) specifier positions. But these facts are also support for an optional juncture in mirror-theoretic word formation in the CP domain quite generally in Bangla. I leave a treatment of these other word order facts and the possibility of a more general juncture point for the restart of the mirror-theoretic word in Bangla for future work.
The thake form of the auxiliary is possible in the non-subjunctive sentence (a) as well, but has an obligatory generic interpretation. In (b), the thake form of the auxiliary is actually ambiguous between the episodic and generic interpretations, and the ačhe is simply impossible.
References
Abels, Klaus, and Ad Neeleman. 2009. Universal 20 without the LCA. In Merging features: Computation, interpretation and acquisition, eds. Anna Gavarró, Josep Brucart, and Jaume Solá, 60–79.
Adger, David. 2007. Three domains of finiteness: A minimalist perspective. In Finiteness: Theoretical and empirical foundations, ed. Irina Nikolaeva, 23–58. Oxford: Oxford University Press.
Adger, David. 2010. Variability and grammatical architecture. LingBuzz: http://ling.auf.net/lingBuzz/001176.
Adger, David, Daniel Harbour, and Laurel Watkins. 2009. Mirrors and microparametric phrase structure beyond free word order. Cambridge: Cambridge University Press.
Baker, Mark. 1985. The Mirror Principle and morphosyntactic explanation. Linguistic Inquiry 16 (3): 373–415.
Bianchi, Valentina. 2003. On finiteness as logophoric anchoring. In Tense and point of view, eds. Jacqueline Guéron and Liliane Tasmovski, 213–246. University of Paris X, Paris.
Biswas, Priyanka. 2013. The role of tense and agreement in the licensing of subjects: Evidence from participial clauses in Bangla. Natural Language & Linguistic Theory. doi:10.1007/s11049-013-9212-x.
Brody, Michael. 2000. Mirror theory: Syntactic representation in perfect syntax. Linguistic Inquiry 31 (1): 29–56.
Brody, Michael, and Anna Szabolcsi. 2003. Overt scope in Hungarian. Syntax 6 (1): 19–51.
Bye, Patrik, and Peter Svenonius. 2011. Exponence, phonology and non-concatenative morphology. Ms., University of Tromsø.
Cinque, Guglielmo. 2004. Deriving Greenberg’s Universal 20 and its exceptions. Ms., University of Venice.
Dasgupta, Probal. 1983. Bangla equatives, complementizers, final foci, and roots. Linguistic Analysis 11 (2): 103–137.
Dasgupta, Probal. 1996. Remarks on subjunctivity. In Papers on language in society (papers in memory of Prof. Srivastava), eds. Shivendra K. Verma and Dilip Singh, Vol. 1. Delhi: Kalinga.
Dasgupta, Probal. 2005. Q-baa and Bangla clause structure. In Yearbook of South Asian languages and linguistics, eds. Rajendra Singh, Probal Dasgupta, Ramakant Agnihotri, Josef Bayer, Peter Edwin Hook, and Tanmoy Bhattacharya, 45–81. Berlin/New York: Mouton de Gruyter.
Davies, William, and Stanley Dubinsky. 2004. The grammar of raising and control: A course in syntactic argumentation. Oxford: Blackwell.
Demirdache, Hamida, and Myriam Uribe-Etxebarria. 2000. The primitives of temporal relations. In Step by step. Essays on minimalist syntax in honour of Howard Lasnik, eds. Roger Martin, David Michaels, and Juan Uriagereka, 157–186. Cambridge: MIT Press.
Enç, Mürvet. 1986. Towards a referential analysis of temporal expressions. Linguistics and Philosophy 9 (4): 405–426.
Enç, Mürvet. 1987. Anchoring conditions for tense. Linguistic Inquiry 18 (4): 633–659.
Giorgi, Alessandra. 2010. Towards a syntax of indexicality. Oxford: Oxford University Press.
Kayne, Richard S. 1994. The antisymmetry of syntax. Cambridge: MIT Press.
Klein, Wolfgang. 1994. Time in language. London and New York: Routledge.
Laka, Itziar. 1990. Negation in syntax: On the nature of functional categories and projections. PhD diss, MIT, Cambridge, MA.
Pollock, Jean Yves. 1989. Verb movement, universal grammar, and the structure of IP. Linguistic Inquiry 20 (3): 365–424.
Ramchand, Gillian. 2005. Two types of negation in Bengali. In Clause structure in South Asian languages, eds. Veneeta Dayal and Anoop Mahajan. Dordrecht: Kluwer.
Rizzi, Luigi. 1997. The fine structure of the left periphery. In Elements of grammar: Handbook in generative syntax, ed. Liliane Haegeman, 281–337. Dordrecht: Kluwer.
Simpson, Andrew, and Saurov Syed. 2013. Finiteness, negation and the directionality of headedness in Bangla. Natural Language & Linguistic Theory. doi:10.1007/s11049-013-9223-7.
Van der Wurff, Wim. 1989. The syntax of participial adjuncts in Eastern Bengali. Journal of Linguistics 25 (2): 372–416.
Zanuttini, Raffaella. 2001. Sentential negation. In The handbook of contemporary syntactic theory, eds. Mark Baltin and Chris Collins, 511–535. Oxford: Blackwell. Chap. 16.
Acknowledgements
I gratefully acknowledge the generous engagement of Probal Dasgupta in writing this paper and for discussing the intricacies of the Bangla data with me. This paper would have also been impossible without the fruitful and ongoing discussions on the topic of Direct Linearization that I have enjoyed with David Adger and Peter Svenonius, both together and separately. The text of the present paper has benefited greatly from feedback from anonymous NLLT reviewers, and from the editors. Thanks also go to audiences at the University of Tromsø and Stuttgart University where this work was presented in informal workshop settings.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ramchand, G. Deriving variable linearization. Nat Lang Linguist Theory 32, 263–282 (2014). https://doi.org/10.1007/s11049-013-9225-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11049-013-9225-5