, Volume 5, Issue 1, pp 1-32

From Old French to the theory of pro-drop

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


The pattern of pro-drop found in OF suggested a revision in the theory of pro-drop. The theory proposed removes the phenomenon from the set of universal parameters and argues for the possibility of reducing it to directional government and feature identification. In this way we begin to explain why some languages do and some do not allow empty subjects. Notice that when one attributes a property to a parameter one is saying, in effect, that that property cannot be explained, at least not linguistically. The main purpose of this paper has been to try to explain linguistically the pro-drop parameter.

The model of pro-drop proposed in this paper has a number of interesting consequences well beyond its ability to account for the facts of OF. Synchronically, it helps explain the distribution of pro within languages and across languages. Diachronically, it helps explain the change from a system which licenses pro to one which does not.

Many questions remain. How and why, for example, did OF acquire V2 effects? Can we simply write this off as a consequence of Frankish-Galloroman bilingualism, or were there dynamics internal to Galloroman itself which made it receptive in some way? We know little about the Gallic (Celtic) substratum, but we do know that the modern Celtic languages have VSO word order. Is it too farfetched then to speculate that this substratum was instrumental in the receptivity of Galloroman to the V2 structure?

Another provoking question is why Romance free inversion should have been unusual in OF. Possibly Romance inversion is incompatible with Germanic inversion and was dispensed with prior to the emergence of written documents. On the other hand, Romance inversion may never have been a property of the Galloroman language which became French. That is, VOS order may have been a development peculiar to Italian and Spanish or to the spoken Latin which became Italian and Spanish. Without Romance inversion, what device did OF use to extract a subject from an embedded clause: optional que, que/que, que/qui? When and how did the que/qui rule of ModF originate? As in English, complementizer and relative que was sometimes omitted in OF. Whether this exactly paralleled the English situation and whether or not the omission was used as a means of proper government is a topic under study.

I owe many thanks. For helpful comments, encouraging words, and the trouble they took, I am grateful to Joe Emonds, Frank Heny, Nina Hyams, Richard Kayne, Paula Kempchinsky, David Lightfoot, Carlos Otero, Carlos Quicoli, Luigi Rizzi, Yves Roberge, Richard Sproat, Robert Stockwell, Barbara Vance, and the NLLT reviewers of this paper. To Tim Stowell, who has seen this project through from its inception, I owe a special thank you. I wish also to acknowledge the many interesting and helpful comments I received from participants at NELS 1985, LSRL XVI, WCCFL V, and GLOW 1986.