VerbCorner: Another Game with Words


Games with Words continues to roll out new games, releasing VerbCorner.

VerbCorner contains a series of sub-games with titles like “Philosophical Zombie Hunter” and questions that sound like they’re straight of an SAT reading comprehension test, starting with a long story I won’t repeat and then questions like:

Michelle smashed the gauble into quaps. Which of the following is *definitely* a real human?

  1. Michelle
  2. None of the above
  3. Can’t tell because in this context ‘smash’ has more than one meaning.
  4. Can’t tell because the sentence is ungrammatical.
  5. Can’t tell because I don’t know that verb.

Josh Hartshorne, one of the developers, writes that Verbcorner is “a new project I’m doing in collaboration with Martha Palmer at CU-Boulder, which is aimed at getting human judgments to inform the semantics assigned to verbs in VerbNet. You can find a press release about the project … You’ll see we’ve incorporated a lot of gamification elements in order to make the task more engaging (and also to improve annotation quality).”

This note from the press release harkens back to my philosophical semantics days, “If we do not want to define words in terms of other words, what should they be defined in terms of? This remains an open research question…”. I’ll say! The project has a generative semantics feel; in the press release, Josh is quoted as saying “There are a few dozen core components of meaning and there are tens of thousands of verbs in English.” Now I’m left hanging as to what a “core component of meaning” is!

2 Responses to “VerbCorner: Another Game with Words”

  1. Joshua Hartshorne Says:

    There are some similarities to Generative Semantics, but we’re mostly following folks like Levin, Pinker & Jackendoff here. I actually know little about GS (well before my time!), so I can’t comment much on the differences. One important thing is that modern modern decomposition theories involve *partial* decomposition: a verb decomposes into a set of abstract semantic primitives that are shared across many verbs PLUS some idiosyncratic, non-propositional information. So “John rolled Mary” means something like “John caused Mary to move in a rolling fashion” where CAUSE, MOVE, and IN_A_X_FASHION are abstract, propositional predicates, whereas “rolling” is represented in perceptual/motor terms rather than as a logical predicate. So you don’t have the problems of the Classical Theory of Concept (e.g., Locke’s), mainly by sidestepping the issue.

    But whether or not this kind of theory is right, most everybody agrees that verbs have meanings, and most everybody agrees that certain aspects of those meanings are more important for syntax, etc., than others (such as causation, movement, etc.). So the data we collect should be useful for anyone interested in verb meaning.

  2. Bob Carpenter Says:

    That’s very interesting. I don’t know much about generative semantics either (it was even before my time in the mid-1980s), but you see it echoing down through Pustevojsky’s lexical semantics and Lakoff and Brown’s approach to metaphor. You went right to the major objection, which is that words have very idiosyncratic meanings and that there are mixed, differential effects on syntax.

    You could take an approach to generating multi-dimensional factors related to meaning that looked very much like a multi-level model in statistics. There’d be global effects, like being a verb, and then varying class-based effects such as the ones you mention (causation, movement), and then idiosyncratic effects. The idiosyncratic effects are then like the noise terms in a stats model that account for whatever isn’t accounted for by the other factors.

    I know there’s a lot of “vector-based semantics” going on. This should let you do things like build latent factor models based on syntactic alternations as well as the more popular topical context. It’d be interesting to see how the latent versions line up with our own preconceptions.

Leave a Reply to Joshua Hartshorne Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s