Make attention to your assumptions


“Science knows it doesn’t know everything, otherwise it would stop.” - Dara O’Briain

Molecular model of proline

"It's only a model". Credit: Peter Murray-Rust via Wikimedia.

In science, we use models to represent some part of the world in which we are interested. Some models are more sophisticated than others, leading to conclusions which are more or less likely to be valid. Because we can’t know about all aspects of the thing we are studying with infinite precision (more on “precision” versus “accuracy” in a future post), we have to assume certain unknown aspects of the subject we are studying.

Pushing this idea to the extreme, the problem of induction (as described by philosophers such as David Hume, one of my favourite thinkers) points out that, as you wander along pondering the nature of reality you cannot , strictly speaking, be totally sure that the ground will still be there when you take your next step.

However, pragmatism and common sense tell us that there really is a world out there waiting to be explored. We have to assume so, or else no conversations are worth having. But we have to be careful with our assumptions.

For instance, when thinking about the problem of induction and assumptions in French (and if the French are famous for anything, surely it’s philosophy), we must be wary of the word assume. Indeed, assumer does not mean “to assume” in this sense at all. Instead it is mostly used in the context of taking responsibility for or admitting something, e.g. “J’assume mes défauts”.

The closest the French and English uses of the word get is in the religious festival of the Assumption (l’Assomption), which celebrates when Mary is supposed (one might even say assumed…) to have been beamed up directly into the sky without passing go. The name of the festival relates to the original Latin root word assumere, which means “to take” or “to remove”, i.e. God simply picked her up. This reinforces the original meaning of “to assume” in the sense of “taking something for granted”, or “taking it as read”.

In French we do have the related word présumer, though again it is not typically used outside the legalistic context of presuming innocence. There is no such word as présomptif, either. For that, there is only présomptueux.

Typically, to say you “assume” or “presume” something in French, you would just say you think it (penser) or that you can imagine it (imaginer). There’s also the slightly old-fashioned estimer, which in turn doesn’t quite mean “to estimate”, although une estimation is “an estimate”. There is, alas, not yet a word for “guesstimate”.

There is also consommer, which means “to consume”. Confusingly, “to consumate a marriage”, is also consommer un marriage. But beware the related, but distinct, consumer, which is closer to “to exhaust”. One could indeed be consumé after having consommé...

So don’t assume that words which seem familiar have the same meaning, or you risk using a flawed model and coming up with the wrong conclusions. Vous êtes avertis. Which, of course, doesn’t not mean “you’ve been averted”, but rather “you’ve been warned”.

Over to you

  • What assumptions have you made about languages which turned out to be wrong?
  • How do you say “to assume” in your language?

Leave a Reply


5 + eight =