Languages with No Future

Whether language shapes the way you see the world has been a hotly contested issue since at least the early 1900s. Now a couple of years ago, Keith Chen, an economist at the UCLA, published a paper that suggested that speakers of languages that lacked strong future tenses tended to be more responsible about planning for the future. The idea is that languages which grammatically separate the future and the present induce less future-oriented behaviour than languages in which speakers can refer to the future by using present tense.

To give you an example of strong and weak future tenses, in German you would say “ich rufe dich morgen an”, using a present-tense verb to express the future action of calling someone, while you would say the same thing in English as “I’ll call you tomorrow”, using the future tense of the verb. German is thus said to have a “weak future-time reference”, while English has a strong FTR. (This is not to say, mind you, that German has no future tense at all. You could just as well use the grammatically perfect form “ich werde dich morgen anrufen”, but usually you’d opt for the above version.)

What Mr Chen found was a negative correlation between the strength of the future tense and attitudes regarding intertemporal issues, such as saving for old age, exercising more and eating less and other such traits that suggest a more responsible attitude towards one’s own future. Speakers of strong-FTR languages tend to be less concerned about these traits, while the opposite holds for speakers of weak-FTR languages.

His findings have drawn a lot of flak from linguists and other pundits, so a chance to test the hypothesis through a natural experiment would be handy indeed. That’s what CESifo Fellows Matthias Sutter and Philipp Lergetporer, with their colleagues Silvia Angerer and Daniela Glätzle-Türtzler, have just done. They report their findings in their latest CESifo Working Paper.

They found a virtually perfect town for their purposes, a place called Meran in South Tyrol, northern Italy. (Of course you know it, but it is worth recalling that South Tyrol was for centuries part of the Austro-Hungarian Empire before being annexed by Italy in the aftermath of World War II. German continues to be, to this day and despite the relocation of Italians from other parts of the country into South Tyrol, the language of nearly 70% of the population.)

Meran is almost too good to be true for the purpose of examining the influence of language on economic behaviour. Of its 38,000 inhabitants, 50.5% speak German and 49.1% Italian (the rest speaks Ladin, originally a vulgar Latin language left over from the Romanized Alps). Social life is fairly segregated, with different media, such as newspapers and TV channels, catering to each group, and different football clubs. Schools are also segregated, teaching in either German or Italian; there are no schools with bilingual teaching. When it comes to area of residence, however, there is almost no segregation, citizens of both language groups living side by side, so that residents face the same living conditions. Both groups are also predominantly catholic.

Our researchers set up a controlled and incentivised experiment in all fourteen primary schools in Meran, encompassing 860 children aged 6 to 11 from monolingual households. In the experiments, the decisions of the children were incentivised with experimental tokens which could be exchanged for such presents as candies, peanuts, stickers, marbles, wristbands and suchlike. The more tokens, the more presents. Each choice problem involved a decision between receiving 2 tokens at the end of the experiment or a larger number of tokens with a delay of 4 weeks.

They found that German-speaking children were significantly more patient in their choices than Italian-speaking children, who were less likely to wait for a later, but larger reward in each single task and in each grade. The degree of patience is the key, since it plays a crucial role in intertemporal choice: it is reflected in investing in education by forgoing immediate earnings from a job, for instance, or in saving for retirement by giving up current consumption and so on.

This general pattern persisted across all age groups, indicating that already at age 6 there is a strong difference between both groups of children.

In order to account for other possible factors influencing the results, they undertook several robustness checks, including the addition of 555 children from bilingual or immigrant households, using a different format to elicit intertemporal preferences, examining the risk attitudes of the two groups concerned, and running a survey among 177 Meran citizens to gain further insights into potential differences in the attitudes of Italian- and German-speaking citizens towards saving and intertemporal choices. The results held every time.

Thus, there appears to be something to Mr Chen’s findings. As counterintuitive as it may sound, thinking and speaking in a more present-tense manner can indeed make you more responsible about your future, by strengthening your self-control.

I’ll start at once. Tomorrow, that is.

Matthias Sutter, Silvia Angerer, Daniela Glätzle-Rützler, and Philipp Lergetporer: The effect of language on economic behavior: Experimental evidence from children's intertemporal choices, CESifo Working Paper No. 5532

Other CESifo Working Papers by Matthias Sutter
Other CESifo Working Papers dealing with language
Other CESifo Working Papers dealing with intertemporal choice