The Mysteries of Time

Linguistics: when the past is in front of us

March 2024

Linguistics: when the past is in front of us

The way we talk about time depends very much on the language we speak: the future can be in front of us or behind us, to the left or to the right, up or down. Here we take a trip through grammar and philosophy.


ime and space are inseparable, as Einstein’s theory of relativity teaches us. They are also inseparable when we communicate: the end of a story is “near”, we have the whole future “in front of us”, and sometimes we have to “turn our backs” on a painful past.

“This association between time and space is found in most languages,” explains Balthasar Bickel, professor of linguistics at the University of Zurich. It is probably linked to the fact that space is less abstract and more immediate than time.

Indo-European languages associate the future with what lies ahead and the past with what lies behind. In this way, we ‘turn’ towards the future and ‘return’ to the past. This spatialisation of time can be observed in two modes.

In the first, passive mode, we remain static and time flows in our direction, from the front. We rejoice at the ‘return’ of spring, which ‘arrives’ after winter, or learn that an appointment has been ‘postponed’. In the second, active, sense of time, we move through time itself, “fast approaching” the holidays and “racing towards the future”.

The meaning of time

We represent time by an arrow going from left to right when we grew up speaking French or German, from right to left for Arabic or Hebrew, and from top to bottom for Mandarin or Korean. There is also a certain amount of diversity in front-back representation. In Vietnamese and Aymara, for example, it is the past rather than the future that is in front of us. In this language spoken in the Andean Altiplano, the word “nayra”, meaning “eye”, is used to indicate both what is in front of us and what belongs to the past. So we say “nayra mara”, literally “before the year”, to mean “the past year”. One interpretation, put forward by certain specialists, is that since the past is much better known than the future, we can see it better in front of us than an uncertain future behind our backs.

Tseltal, a form of Mayan spoken in Mexico, sees the future as above and the past as below. The same is true of the language of the Yupno people of Papua, whose concept of time is associated with a topography marked by numerous hills. Time follows the trajectory of the sun in Pormpuraaw, an Australian Aboriginal language, placing the past in the east and the future in the west. These multiple dimensions can also be mixed: people speaking a Moroccan dialect of Arabic associate the future verbally with the front, but gesturally with the back.

I speak, therefore I think

Balthasar Bickel continues: “Since we don’t have the gift of telepathy, we construct a large part of our culture through words. The way we talk about time influences how we perceive it and think about it.” Associations between time and space are deeply rooted in our brains: experiments show that we take longer to classify a word such as “yesterday” in the “past” category if it appears on the right of a screen than if it appears on the left, for the simple reason that the arrow of time runs, for us, from left to right. The opposite is true for Arabic speakers.

We could therefore assume that linguistic differences fundamentally change the way we conceptualise the world. This is the essence of the Sapir-Whorf hypothesis, formulated in the 1930s and inspired by the concept of relativity in physics. But “there are many traits of culture that develop and are transmitted without the spoken word, for example ritual movements,” notes Balthasar Bickel. Until the 1990s, the Sapir-Whorf hypothesis remained primarily a theoretical and philosophical question. But since then, empirical research in psycholinguistics has put it to the test, and confirmed it for most domains”.

Language influences our thinking, but does not determine it entirely, notes Jürgen Bohnemeyer, professor at the University of Buffalo: “I think that cognition has an innate basis that remains little influenced by language, particularly when it comes to fundamental things like grasping the temporal order of events”.

Time without time

Jürgen Bohnemeyer’s doctoral work on the Yucatec Mayan language, published in 1998, helped to challenge Sapir-Whorf’s linguistic relativism. Yucatec does not have a verb form that distinguishes between the future, present and past, explains the linguist. This characteristic is not in itself unusual: it applies to around 40% of idioms around the world, particularly Mesoamerican languages. The same verb form, such as “I eat”, can therefore mean the present as well as the future or the past. Most languages can then use a temporal connector (“since, while, before, after…”).

Yucatec is different: it has no temporal connector. It cannot express the future or the past directly through a conjugation or tense or with the help of connectors. “When I was doing my doctorate, I wondered whether this might reduce the ability to describe sequences of events”, explains Jürgen Bohnemeyer. To answer this question, he set up an experiment. The first person watched two videos showing the same events (drinking water, throwing a ball, etc.), but in a different order. By asking closed questions (requiring a “yes/no” response), he had to guess which of the two videos had been watched by a second participant.

The linguist then observed that the success rate with a pair of Yucatec speaker was the same as when the participants spoke German. This seminal work shows that the absence of temporal markers in Yucatec in no way prevents the ability to clearly communicate the temporal order of events - and thus tempers Sapir-Whorf’s cultural relativism.

So how is it possible to communicate indications of temporality without having temporal markers? Jürgen Bohnemeyer’s doctoral work provides the answer: it’s to be found in the linguistic concept of ’aspect’, which describes whether an action is finished, in progress or repetitive. French easily makes this kind of distinction in the past tense (“je répondais, je répondis”), English in the present tense (“I am reading, I read”), while German needs auxiliary words such as “plötzlich”, “als”or “immer”. With a grammar that is here much richer than that of Indo-European languages, Yucatec uses verb forms to describe notions of aspect directly, including the fact of being about to carry out an action or of having just completed another.

With his work, Jürgen Bohnemeyer has explained how the combination of aspects - the fact, for example, that one action has been completed while another has just begun - makes it possible to communicate the temporal order followed by events. “It’s a mistake to think that a language cannot function without certain grammatical forms,” says the linguist. Some idioms are gendered, others are not, without preventing distinguishing between the sexes. The reason is that we are constantly inferring and completing what has not been explicitly stated".

The researcher underlines our cultural biases: “Our European-centricity leads us to be quick to point out the deficits of other cultures, by saying, for example, that Mayan languages lack temporal markers. But it’s also important to reverse the perspective and note that our Indo-European languages lack the verbal forms of aspect that make Mesoamerican languages so rich.”

Language influences the way we see the world, and the reverse is also true. Linguistic developments are clearly linked to social changes, such as the masculinisation of the French language in the 17th century and its feminisation since the end of the 20th century. Are the specific temporal grammatical structures of a language such as Yucatecan linked to the environmental or social conditions of the people who speak it? Jürgen Bohnemeyer: “This is a difficult question to answer. The absence of verbal temporal markers is widespread around the world, and is found in very different societies”. Balthasar Bickel of the University of Zurich agrees: “It is impossible to draw conclusions from a single case. Testing such a relationship would require worldwide databases on nonlinguistic facts and advanced statistical models, which we are working on.”

The Aztec sun stone, dating from the early 16th century, is a monolith that represents the Aztec cosmovision of time. Housed in Mexico City's National Museum of Anthropology, it measures 3.6 metres in diameter and weighs over 24 tonnes.
The Aztec sun stone, dating from the early 16th century, is a monolith that represents the Aztec cosmovision of time. Housed in Mexico City’s National Museum of Anthropology, it measures 3.6 metres in diameter and weighs over 24 tonnes.

Life without duration

Linguistic differences in the way time is described can go even further. The ethnolinguist Vera da Silva is studying the idioms spoken by three indigenous peoples of Brazil, the Huni Kuĩ, Awetý and Kamaiurá, who number only around ten thousand speakers in all. Although these cultures are hundreds or even thousands of kilometres apart, they have one thing in common: they have no metric concepts of time, i.e. words that refer to durations.

Like Mesoamerican languages, these idioms have a wealth of markers of aspect and modality, distinguishing, for example, between the attested past (which really happened) and a reported past, or between the future that is certainly going to happen and the future we hope for.

“In Western society, time is often considered through the prism of duration, such as weeks or hours, holidays or a morning,” explains the researcher, now at the University of Bergen in Norway. “The cultures I study, on the other hand, have a time based on events. It could be an important stage in life, sunrise, a meal eaten together or hunting. People never talk about how long an event might last, but rather consider whether the next event will have started after or before the end of the first. So there is no single chronology, no temporal axis on which to anchor an event precisely.”

The researcher gives a striking example: “In 2015, I visited a Huni Kuĩ village in the state of Acre, near the Peruvian border. My contact was reluctant to tell me how long the river journey might take. At my insistence, he finally said: ‘About a day’. In the end, the trip took a week: the current was particularly strong, the engine had a problem, we helped some communities along the way... But the delay was no problem for the villagers, who weren’t expecting me at any particular time. It shows that in an environment like this, thinking in terms of time isn’t necessarily very useful”.

Establishing a relationship of trust is absolutely crucial for this type of fieldwork, continues Vera da Silva. “I had to forgo certain questions, because the villagers prefer not to express themselves individually but rather collectively, to act as representatives of their culture. The experiences I imagine are rooted in the way I conceptualise time, and I always have to remember to question these cultural biases. In our society, time as a duration has a certain value, we are chronically short of it, and time is, as the saying goes, money. Working with these communities is incredibly enriching for me, because I then try to enter into their temporal framework: a time that is less counted and closer to the simple fact of being.”

The Europa Star Newsletter