LISSIM 6

June 1-15, 2012@ Kangra

Selected Essays

An exploration of “go” as a light verb in English

Anne Beshears
MPhil, JNU

How would LISSIM 6 help you in sharpening your research question and in looking for a possible answer?

One of the reasons that I would like to attend LISSIM 6 is because I would like to explore as many areas of syntax and semantics as possible before formally deciding on a research topic. While I am still in the early stages of deciding, exploring as many aspects of syntax and semantics as I can before settling on any one thing, I have been doing some work in complex predicates which has led to some interesting questions for further consideration.

In my paper, “To go and V”: an exploration of “go” as a light verb in English, I have attempted to describe what is happening in the construction, “to go and V,” in Southern American English. This construction is highly productive and connates an action which is complete and (generally) at which the speaker is expressing surprise or dismay. For example, “Jim went and did something stupid.” In these constructions, there is no sense of movement or going but simply a completed, purposeful action. Like other complex predicates, this is a serial verb-like construction and yet represents only one, not two events

Often, which of the readings is intended is determined by context and inflection; determining which of the two readings is at play often relies solely on speaker intuition. In describing these construction, I have attempted to give some diagnostics to show that the single eventive reading does indeed exist and to disambiguate it from the verb sequenced reading. This same type of construction, with two tensed verbs joined by conjunction acting as a single eventive predicate, can be found in a variety of other languages as well, including Indo-European languages such as Danish and Swedish as well as South Asian languages including most Dravidian languages (Malayalam, Telegu, and Tamil for example) and the Dravidian influenced Indo-Aryan languages Bengali, Assamese and Marathi.  As Miriam Butt states in her 2003 paper (pg. 6), “[M]onoclausality can be established conclusively, but … must be done so on a language internal basis. … The identification of complex predicates and light verbs therefore presupposes a very careful scrutiny of the syntax of a given language.”

If the single event reading is distinct from the verb sequenced reading, then it follows that each of the two readings has a distinct syntactic structure. The single event reading must involved two inflected verbs, joined by conjunction, with a single verb phrase. I explore several options for the syntactic structure of this construction, showing that VP/vP shells, both in the X-bar and minimalist frameworks, are insufficient for capturing the distinctive nature of this structure. Alternatively, I propose a possible analysis incorporating Barbara Citko’s parallel merge (Citko 2005:480). This type of construction simply and elegantly solves the issues found in a VP/vP shell analysis and, additionally, lead to some interesting implications.

In my analysis, the “go” light verb, found in little-v, lacks the sense of DIR yet retains the features of GOAL and PURPOSE (or requiring an agent). The VP becomes the metaphorical space and is taken by the vP as the GOAL. But, a significant complication in these cases is the presence of the conjunction and the fact that both verbs take tense. Using Citko’s parallel merge analysis, this is resolved by having two vPs, both taking the VP as a sister-complement, and conjoined by a conjunction. Thus, the tense may lower (in English) onto the adjoined little v, giving both tense. (See Figure 1 below.)

There are also cases in English where the “go” light verb may get tense while the lower (or second) verb does not. In these cases, it means that the action was attempted but not completed. In this case, the “go” light verb still takes the VP as a goal[1] but there is no second v as the event did not happen. For example, in (1) below, the speaker attempted to leave for the party (+PURPOSE, +GOAL) but the leaving event (the goal) was not completed.

    1. I went to leave for your party but, just as I was about to walk out the door, I got an urgent phone call.

 These same variations can be found in Malayalam. Sabu M. notes in his MPhil dissertation that in Malayalam, a Dravidian language with as many as ten light verbs, there are two light verbs which can be used to express the “undesirability of the occuring of an event that is contrary to the expectations of the speaker.” (Sabu M. 2008:24)  These are the light verbs analogous ‘to go” and “to discard” He gives the following examples where example 2. means that “Ram was frightened” and 3. expresses the meaning that “Ram was frightened” and this was unexpected or undesirable (Sabu M. 2008:25). This would be roughly analogous to an English sentence “Ram went and got scared (over nothing).”

    2. ram peedichu

       ram frighten-pst

       Ram got frightened.

   3. ram peedichu     pooyi

       ram frighten-pst go-pst

       Ram got frightened.  (or: Ram went and got scared.)

 The parallel merge construction seems to most adequately encapsulate the intuitions of what is happening in the construction, but it is not without its problems. Despite its elegance, it fails when taking a feature-based or probes and goals based approach. Similarly, Citko uses an asymetric conjunction as the motivation for linearization, but this is actually unmotivated by the system and uninterpretable. I propose that a symetric conjunction is necessary, where the asymetry must be motivated by the relationship between the two vPs rather than the conjunction itself.

While parallel merge appears to resolve many of the issues found in other structures, there are some significant problems when taking a minimalist approach which must be addressed. Specifically, if parallel merge is possible, then this leads to some interesting dilemmas when discussing interpretable/uninterpretable features or probes and goals. These are things which must be explored further, not only in English but in the other languages which appear to incorporate this same type of construction.

 

 

 

 

For enquiries, write to secretary@fosssil.in