On Booze and Language of Thought

Last Sunday, I was having a discussion with my mother about my drinking – which has been sporadic at best and non-existent at worst. She said she had a probelm with even my sporadic ingestion of alcohol, and demanded that I completely give up drinking. I tried my best to draw it away from a religious/emotional argument, and tried to draw her into a logical argument.

My mother is a biologist by training (it is another matter that her career was in accounting) and said that she is concerned about her gene, and that given that I’m her only offspring, she naturally has incentive in my offspring, and she wants to make sure that they’ll be in good health, and live well. She also has this notion that if either of the parents drink, the children will be born dumb, and there is an increased risk of abnormality. I have no clue about these matters, and somehow managed to change the line of argument.

Then she released her brahmastra. Or what she thought was her brahmastra. Everyone she know who drank alcohol, she said, had ended up becoming a drunkard and a wifebeater. She gave examples of classmates from school, colleagues, colleagues’ husbands, relatives, etc. It was evident that she had prepared her argument well. And in each of these cases, there was no doubt that the person in question was a drunkard and a wifebeater, whose kids were most likely to end up as losers.

It is pertinent to point out here that the entire argument was happening in Kannada. In fact, I’ve never talked to either of my parents, or to any other close relatives, in any other language. I am in general fairly fluent at the language, at least at the Bangalore version which includes loads of English words. I have in general managed to hold my own in several debates and discussions at social gatherings, while talking exclusively in Kannada. I have explained to relatives complicated financial products, and how the sub-prime crisis unraveled, all in Kannada.

Getting back to the argument, the best way for me to handle my mother’s examples was to provide counterexamples. Of perfectly decent people who consumed alcohol. For statistical reasons and given the way the hopothesis had gotten framed, I would need a much larger list than my mother had produced. And in the spur of the moment, I decided that I wouldn’t do a good job at listing and I should continue with logic-based arguments. Phrases such as “selection bias” and “Bayes’ theorem” and “one-way implications” flashed across my head.

Holding up your end of the debate when you aer talking nomally, and in Kannada, is fair enough. However, when you are extremely animated, and speaking at 100 words per minute, things are a bit harder. I realized that my mother would understand none of the jargon that had flashed across my head, and I’d have to explain to her in normal language. My mouth was processing words at 100 words per minute, and suddenly my brain seemed to have gotten a bit slower. The pipeline became empty for a moment and i started stammering. And my mother started making fun of my stammering (i used to stammer a lot when I was a kid. took a lot of effort to get over it).

Coming to the crux of this essay – at this moment another thought flashed to my head. For a long time I wasn’t sure if I thought in any specific language, or if my thought was general. And even if I thought in a particular language, I wasn’t sure if I thought in Kannada or in English. I had always done well enough in both languages to keep this debate unresolved. Now there was the data point. The clinching data point. I quickly realized that had I been speaking in English, my pipeline wouldn’t have gone empty. In fact, when I was trying to explain stuff to my mom, I was doing two levels of translation – I was first translating jargon into normal English, and then translating that into Kannadal. Powerful evidence to sugggest that I think in English.

I was so kicked by this discovery that the original argument didn’t matter to me any more. I quickly promised my mother that I will never consume alcohol again, and she said “shiom”, a kid-word that means something like “ok what you’ve said is final and binding and no changing it”. So I suppose that is how things will stay. I will henceforth stop consuming alcohol. Not that I’ve been consuming much nowadays – average one drink every two months or so. It won’t be hard at all to make the transition.

PS: Interestingly, when I’m trying to speak in any Indic langauge (Hindi, Tamil, etc.) I instinctively form my thoughts in Kannada and then translate. Maybe it is because of similarities that the cost of translation from Kannada to these languages is much lower than the cost of translation from English, that it becomes profitable for me to think in Kannada, which is harder than to think in English.

25 thoughts on “On Booze and Language of Thought”

    1. no i think it depends on the extent to which we use english. if in your normal daily life you tend to use some other language much more, then you will automatically start thinking in that.

  1. Really nice post!!

    Could it also be that in this case your mind was unsure whether to present a logical answer or an emotional answer..and the language translation problem is irrelevant? Would you have held your corner if you and your mom were arguing about the drinking habits of a third person?

    1. no no no. there was no such dilemma. i’m sure that I was doing a logical argument only, and was trying hard to translate Bayes’ theorem, selection bias, etc. into normal kannada.

      nowadays, during such discussions, i try to make things impersonal. detach myself from my argument. so this won’t matter

  2. As a drinker, am I (in your mother’s head) destined to become a drunkard and wife-beater? Just thought that was interesting… the automatic link between alcohol and men.

    The (only) good thing about only really speaking one language fluently is that the question of what language you think in doesn’t arise.

  3. Very interesting post..

    Maybe, we tend to think in the language we read more, rather than speak more. After all written matter is someone’s expressed thoughts

    1. or maybe it is the total amount of communication, appropriately weighted by the complexity of thought in each language that matters.

      which fits in with your point – what we read is probably more complex than what we speak at home everyday.

      1. Maybe it is also a case of the breadth of vocabulary in a language….I know many more words in English to express feelings at all levels, intensity, complexity etc. and yet when it comes to expressing relationships, I find Kannada richer – with separate words for each. There is no way I can find a good word for “voragithi” in English and so, if I am speaking to a Kannada+English speaker, however complex the thought, I will use the word “voragithi” instead of saying “my husband’s brother’s wife” not even “brother-in-law’s wife” as a brother-in-law is my sister’s husband…

        1. yes – this is also very possible. just like there is no easy way to say voragithi in english, there is no easy way to explain bayes’ theorem, selection bias, etc. in kannada. so this might also have been the reason that i struggled.

  4. Really. You cannot think in either Kannada or English. Your brain processes ideas using a representation that has to be different from the languages that we speak to prevent a bottomless recursion (also the representation the brain itself uses needs to be unambiguous, unlike human language).

    All that is happening is that your brain is faster at transforming certain domains of thought into English than into Kannada.

      1. haven’t read that hypothesis, but this is my theory: we store an abstract representation of an idea or a concept separate from the word for that and link the two. Take the colour red. The representation of that concept will be what it looks like. If we learn that colour in kannada first, then we associate the word ‘kempu’ with that concept. Later if we learn that that colour is called ‘red’ in English, then we just associate red with kempu. So when we see the colour and have to express it in English, we first think kempu and then translate it as red. Of course, over time if we find that we are using red more than kempu, then we just build a direct association of red with the colour and bypass kempu.

        I am sure for some of the concepts you learned in kannada first and don’t use much in English, you will still be thinking those concepts in Kannada. Since you learnt Bayes’ theorem only in English, it is not surprising that you have to think it first in English and then translate to kannada.

        1. fair logic. what you say makes sense. but again – the kind of stuff i’ve written here – there can be several valid explanations to that which are not contradictory. and yours is definitely one of them.

  5. Interesting that sum1 posted this.. i have thought abt this myself many time and reached the conclusion that I think in my mother tongue (gujarati).. odd given that I only speak the language at home and virutally nowhere else outside.
    keen to know if this is true for others as well

    1. for a long time, i too used to think that i think in kannada. but this particular incident contradicts that. but maybe it’s context-sensitive. if someone were to wake you up in the middle of the night and ask you to explain Bayes theorem, i’m sure you’d start off in english.

  6. By forgetting what the original argument was and getting excited by a trivial ‘what language do you think in’ question – you have committed adultery to the (in)famous red bandana.

    Today – it must be twisting and turning in its little grave

Leave a Reply to skimpy Cancel reply