Language Log: Can grammar win elections?

That's the title of a recent paper by Caitlin Fausey and Teenie Matlock that appeared in the journal Political Psychology. It's a heartwarming title, one that permits me to dare to dream of that better day when political parties will divert rivers of cash to linguistics departments, when a grad student will be able to defend a thesis on applicative constructions in East Asian languages one day and take up a lucrative job as Washington policy wonk the next, and when volumes by Noam Chomsky and Richard Montague will be pressed into the hands of military personnel charged with the task of winning the hearts and minds of residents in troublesome, volatile nations. The paper stems from recent interest in the persuasion sciences about the fact that how a message is expressed often has a startling impact on the choices and behaviors of its audience. Most of the attention has been lavished on questions of lexical choice, or on whether a message is framed as involving gains rather than losses. But these are happy days, and persuasion research seems to be taking a more adventurous turn, with investigators beginning to tackle questions involving finer points of semantics and their grammatical correlates. For instance, David Beaver recently posted here about a paper showing that, when people were asked to answer "How important is it to you to be a voter in the upcoming election?" they were quite a bit more likely to actually cast their ballot in the election than if they were asked "How important is it to you to vote in the upcoming election?" In the current brazenly-titled paper by Fausey and Matlock, the finding was that subjects expressed greater skepticism about political candidates when the politicians' sleazy actions were marked for imperfective aspect ("X was taking hush money") rather than for perfect aspect ("X took hush money"). The inspiration for the study, according to the authors, comes from work on the psycholinguistics of verb phrases and event representations. The linguistic notion of aspect captures the idea that in describing events and actions, we often encode information not just about the nature of an event or action, and whether it's located in the past, present or future, but also something about which component of the event is being described. For instance, the imperfective aspect (was building a house) emphasizes the ongoing nature of the action or event, while the perfect aspect (built a house) puts focus on its result or end state. Different aspectual classes seem to change the psychological prominence of elements in an event. It so happens, for example, that the word hammer is more readily primed (i.e. subjects take less time to recognize the word) upon reading a sentence such as "He was pounding the nail" than upon reading "He pounded the nail" (Truitt & Zwaan, 1997)—this suggests that the concept of a hammer was more active in people's minds while reading the imperfective version. Similarly, Ferretti et al. (2007) discovered that an isolated phrase (was skating) primed a typical location of the event (arena) but only when the verb was marked as imperfective, and not in its past perfect form (had skated). On the basis of their own results on the effects of grammatical aspect on political opinions, Fausey and Matlick draw some general inferences about how imperfective aspect serves to highlight negative actions, and conclude that "grammar can influence electability." Fausey and Matlick's study is intriguing, and I feel a warm flush of pleasure whenever psycholinguistic evidence breaks beyond the tight circle of language geeks who gather it, and finds a broader audience—I truly do. But before I go and advise political candidates to sprinkle some imperfective morphology throughout their attack ads against opposing candidates in order to make their indiscretions loom larger than life, I'd like to figure out exactly what it is I'm entitled to conclude from the study. To give up some of the details of the experiment (there were actually two experiments reported, but I think that similar issues might arise for both): Each subject read a single short paragraph describing a fictional candidate. For example, one version of the description read as follows: Mark Johnson is a Senator in the United States Senate. He is up for reelection. He graduated from the University of Texas, Austin with a degree in political science. Mark's first term as a United States Senator is almost complete. Last year, Mark had an affair with his assistant and took hush money from a prominent constituent. An alternative version translated the critical last sentence into its corresponding imperfective form: Last year, Mark was having an affair with his assistant and was taking hush money from a prominent constituent. There was also an orthogonal manipulation which resulted in versions in which the actions described in the last sentence were positive, as follows: Last year, Mark rekindled his relationship with his wife and collected donation money for the American Cancer Society./ Last year, Mark was rekindling his relationship with his wife and was collecting donation money for the American Cancer Society. After reading whichever version of the paragraph they'd received, subjects answered the following questions: a) Will this candidate be reelected? b) How confident are you about your decision regarding reelection? (using a 7-point scale) c) Please estimate the total amount of hush money/donation money For negative actions, subjects responded more often and with greater certainty that the candidate would not be reelected when the actions were marked for imperfective aspect. They also guessed the amount of hush money to be higher in the imperfective condition. When it came to the effects of aspect in describing positive actions, however, aspectual marking made no difference whatsoever. The authors deal with this discrepancy between negative and positive actions by pointing to a tidy pile of studies showing that negative information is especially powerful in shaping attitudes or choices—perhaps, they propose, the more generally robust effects of negative information provided enough space within which to observe an impact of aspectual marking. Maybe. But I'm guessing that this doesn't fully explain the now-you-see-it-now-you-don't effects of aspect. I suspect some of it comes from the fact that there really isn't a tidy relationship between grammatical aspectual markers and their psychological effects, because such markers are only one cue among many to the structure of events. In other words, I'd be willing to bet good money that blindly following a general rule of slapping imperfective markers onto an opponent's dubious actions would be about as effective a persuasive strategy as always using a passive construction to minimize your own errors—that is, useful in some situations, but irrelevant or even backfiring in others. More specifically: Imperfective morphology is actually ambiguous between two kinds of event structures. It can mark events as being in progress. For example: I was holding my breath; or I was driving home; or I was building a house. These phrases represent events or activities that were ongoing at some particular time in the past—the events may or may not have ever come to a satisfactory conclusion. For instance, I may never have made it home, and the house may never have gotten built. But the imperfective can also be used to express events that happened repeatedly or even habitually. For example: I was taking art classes, I was using foul language, I was dating that morose painter, all of which suggest events that successfully happened more than once. So, imperfective morphology maps onto more than just one kind of event. The direct objects themselves often provide a clue as to whether the event is progressive or habitual, but not always—sometimes extra information in the sentence or the general context is needed for clarification. For instance, drinking beer can be understood either as progressive or habitual: I was drinking beer in the bar when my sworn enemy crashed through the door. I was drinking beer in those days, but I've since moved on to hard liquor. Getting back to Fausey and Matlick's paper. I think this is important, because my sense is that the positive and negative versions of the experimental items may actually lend themselves to different readings of the imperfective phrases. For instance, rekindling a relationship is something that normally takes some time, so it's easy to see this as an event in progress that may or may not have had a successful outcome: Last year, I was rekindling my relationship with my spouse, but we never quite managed to succeed. But it's a bit harder to get this reading for having an affair—alas, perhaps due to the fact that infidelity takes much less time and effort to accomplish than renewing a marriage. It seems decidedly odd to say: Last year, I was having an affair with my assistant, but we never quite managed to succeed. (I think the contrast is less sharp in the case of collecting donation money versus taking hush money, but still possibly there—at the very least, I'd want to empirically satisfy myself that this ambiguity has been controlled for.) This means that the observed effects of aspectual morphology on attitudes could have been different for positive versus negative actions not because the imperfective generally makes the activity itself more salient (though it might), but because very different inferences are involved in the specific examples that were used. The phrase having an affair as compared with had an affair suggests that the illicit sexual activity happened more than once—an amplification of the bad behavior. But when you compare was rekindling his relationship with his wife to rekindled his relationship with his wife, the imperfective could well be attenuating the positive behavior rather than amplifying it, because it leaves open the possibility that the attempt failed. So a note to political strategists and speechwriters: Sorry. A quick and dirty grammatical rule of thumb is unlikely to do the trick. Orchestrating the desired effects in the minds of voters may turn out to involve, unfortunately, more nuanced efforts. You'll likely have to hire a trained semanticist to your campaign team. And do some more studies. Might I suggest an increase in funding to linguistics departments?