The past is before us, or behind, whatever

A recent paper in Cognitive Science by Nuñez and Sweetster has evoked several interesting strains of blog commentary. The paper is about the cognition of time as a function of conceptual metaphor with space.

That's pretty abtruse-sounding, and the Wikipedia entry on conceptual metaphor is fairly informative. The basic idea is that we may understand one class of relationships (a conceptual domain) in terms of the relationships that we know apply to another analogous domain.

All the talk about the new article seems to have come from the attention from a New York Times article, which is imprisoned behind TimesSelect, so I'll quote the interesting Savage Minds post instead:

The New York Times is running an article on a recent article in Cognitive Science by Nunez and Sweetser which demonstrates that Aymara speakers imagine the past to be in front of them and the future behind them -- reversed, in other words, from the spatial metaphors we use in English. The Times article notes "If they are right, this is bigger than anything the 60's tossed up. Is it possible that human concepts of time can vary this much because of language and culture? And what would it be like to think this way? Do I have the rest of my life behind me? And how can I let bygones be bygones if they're right in front of me?" Nunez and Sweetser also makes a to-do about the rarity of this pattern, since, it claims that "so far all documented languages appear to share a spatial metaphor mapping future events onto spatial locations in front of Ego and past events onto locations behind Ego."

The flavor of the SM comment is that cognitive scientists often ignore cultural variability that anthropologists hold as common knowledge:

Cognitive Science produce attention-grabbing headlines much more frequently than anthropologists, and this article is a prime example of how they manage to do so: ignorance.
Have Nunez and Sweetser actually conducted some sort of exhaustive examination of 'all documented languages'? No. In fact their citations reveal that they have examined a grand total of seven: English, Wolof, Chagga, Chinese, Japanese, Turkish, and American Sign Language (to be fair one of the articles they site has 'more cross cultural data').
If Nunez and Sweetser had looked a little bit further -- for example to the Pacific -- they would have found that these sorts of metaphors are quite common.

After the original post, there is a discussion in the comments at Savage Minds with links to elsewhere, including a Language Log entry on the paper:

I feel a need to address recent controversy regarding the uniqueness of the Aymara conceptualization of time-as-space. I cannot respond to everyone who says that their language of choice also has a "back to the future" metaphor, nor will I attempt to reconstruct all of the linguistic (metaphor-based) arguments involved. However, many of the objections that I have heard (and that I am sure the researchers of Aymara asked themselves) are based on a misconception that if a language has a single word that is polysemous between "front/past" or "back/future", then it automatically makes Aymara non-unique.

The short post then discusses why the Aymara case may be different from many others, which centers on the use of gesture as another communicative mode that redundantly includes the front:back::past:future axis.

Chris at Mixing Memory gives some commentary on the entire subject:

I've said it before, and I'll say it again: conceptual metaphor theory sucks. Why does it suck? Well, because there's no experimental evidence for it (and plenty of evidence against it). Except, that is, in one domain: time. Specifically, the work of Lera Boroditsky, along with Dedre Gentner and her colleagues, has provided interesting demonstrations of the influence of the way we talk about space on the way we conceptualize time. I've talked about their work before, and now Dave's talking about Gentner's work over at Cognitive Daily, so I won't go into a lot of detail. Instead, I'll give you an idea of what's going on with the time-space metaphors in their work, and then discuss some recent work by Rafael Nuez and his colleagues which introduces new types of time-space metaphors. The conclusion generally drawn from this work is that time is conceptualized metaphorically through mappings onto space. At the end of this post, I'm going to argue that no current evidence actually supports that position.

The critique involves the troublesome problems of irrelevant meanings and priming effects -- essentially, although languages may be constructed by applying metaphoric meanings to words, there is little evidence that the mind constructs concepts using these metaphors, and testing the cognitive treatment is very difficult considering the linguistic entanglements.

I don't particularly have any opinion, but it has been interesting reading much of these exchanges, which illuminate one present-day aspect of the Sapir-Whorf language-shaping-cognition paradigm.


Nuñez RE, Sweetser E. 2006. With the future behind them: Convergent evidence from Aymara language and gesture in the crosslinguistic comparison of spatial construals of time. Cognitive Science 30:1-49. Abstract