Decomposing French questions

Cedric Boeckx
University of Connecticut

In this study, I propose an alternative account of French wh-in situ constructions based in part on previously unnoted facts concerning the interpretation of in-situ wh's.

Many-widely diverging-accounts exist of the 'problematic' optionality French exhibits in question formation (fronting (1) or in-situ (2)), but no one discusses, and as far as I know predicts interpretive differences between the two strategies, and yet such differences exist. Thus, while it is perfectly possible to answer a 'fronting' question by 'nothing,' it is not possible to do so in the case the wh has remained in situ (3). In that respect, wh in situ patterns like clefting (4). I suggest that wh's in situ, like clefted elements, are focused (i.e., they correspond to detail-information seeking strategies, which differ from both general information seeking (achieved via wh-fronting) and echo questions). The analysis of focus I will develop for French capitalizes on the results of Zubizarreta 1998. Zubizarreta shows how French is unique among Romance and Germanic in the way it realizes focus, and claims that French is in a transitional stage of language change, which accounts for why the language makes use of two distinct strategies to solve conflict between the Nuclear Stress Rule, the Focus Prominence Rule, and word order, viz. node invisibility or Prosodically-motivated movement. These two strategies, I argue, correspond to the fronting and in-situ strategies for question formation.

A question now arises as to whether a (focused) wh-in situ moves at LF. Rooth 1985 convincingly argues in favor of in-situ interpretation of focus. I will follow him in that respect and claim that a wh in-situ stays in situ throughout the derivation (cf. also Reinhart 1995, Tsai 1994), but that something else is moving to C, viz. (the question particle) Q (cf. Hagstrom's analysis of wh in situ in Japanese, where movement of Q is motivated). This movement operation, I argue, is necessary to account for Beck effects with wh in-situ in French. Beck effects (Beck 1996, Pesetsky 1998) arise when a quantifier stands on the way of the wh in situ to C (5). Following Chang 1997 I observe that a quantifier forces a wh in situ to receive an echo (i.e., not truly interrogative) reading (6). I suggest that Beck effects are to be analyzed as standard Relativized Minimality effects, where the quantifier stands on the way to C, and prevents movement of Q. Q-raising being blocked, the sentence cannot be interpreted as a true question (see Hagstrom 1998 on the role of Q in question interpretation). I then examine multiple questions, and D-linked in-situ wh's, and conclude that Beck effects are absent in those cases (7), which leads me to propose a novel analysis of D-linking supporting Pesetsky's 1987 non-movement analysis. For multiple questions, I extend Richards' Principle of Minimal Compliance (8).

In sum, this study draws partly on new data to propose a new view on wh in situ in French. The analysis supports Zubizarreta's 1998 account of focus in French, Hagstrom's 1998 Q-movement hypothesis, and also the Chengian view (Cheng 1991, Tsai 1994) on the typology of wh-questions, which claims that apparent optionality in wh strategies must result in interpretive differences. Interestingly, Cheng shows that Egyptian Arabic, Bahasa Indonesian, and Palauan front wh's in a cleft-like fashion, otherwise they leave them in-situ. In this study I show that French is the mirror image of these languages in that cleft and in-situ strategies go hand in hand-a novel aspect of parametric variation in question formation. In providing evidence for the Chengian view, this study can also be seen as lending further credence to the 'minimalist' claim that genuine optionality does not exist in human languages.

  1. Qu'as tu achetÈ?

What-have you bought

  1. Tu as achetÈ quoi?

You have bought what

  1. Answer to (1): T 'Rien' ('nothing')

Answer to (2): * 'Rien'

  1. C'est quoi que tu as achetÈ?

It is what that you have bought

Answer: * 'Rien'

  1. Beck effect

A semantic restriction on a quantifier (including wh) may not be separated from that quantifier by a scope-bearing element

Quantified structures as barriers for movement

*[ ... Xi ... [Q ... [... ti ...]]]

  1. *Tous les enfants ont lu quoi? (* as non echo)

*[Ci ... [Tous ... [ ... wh-Qinterrog, particle i ]]]

  1. a. Tous les enfants ont lu quel livre/lequel de ces livres? D-linked wh's

All the children have read which book/which of these books

b. Qui a donnÈ tous les bonbons ‡ qui? Multiple question

Who has given all the candies to who

  1. Principle of Minimal Compliance (Richards 1997)

For any dependency D that obeys constraint C, any elements that are relevant for determining whether D obeys C can be ignored for the rest of the derivation for purposes of determining whether any other dependency D' obeys C


Beck, Sigrid. 1996. Quantified structures barriers movement. NALS Chang, Lisa. 1997. situ phenomena French. thesis, University British Columbia.
Cheng, Lisa. 1991. typology questions. PhD, MIT. Hagstrom, Paul. 1998 Decomposing questions. PhD, MIT.
Pesetsky David. 1987. situ: movement unselctive binding. representation (in)definiteness. Press. Pesetsky David. 1998. Phrasal movement kin. Ms., MIT.
Reinhart Tanya. 1995. Interface strategies. Ms., OTS.
Richards, Norvin 1997. What moves where when which language. PhD, MIT.
Rooth Mats. 1985. Association eith focus. PhD, UMass Amherst.
Saito Mamoru 1998. binding unselective binding. Talk given MIT. &0"%
Tsai Dylan. 1994. economizing theory dependencies. PhD, MIT.
Zubizarreta Maria Luisa. 1998. Focus, prosody, word order. Press.

< Back to List of Abstracts | Back to PLC23 Home Page >

About the PLC23 Committee
Previously held Penn Linguistics Colloquium: PLC22 (1998), PLC21 (1997)

Penn Department of Linguistics
University of Pennsylvania