The Whorfian hypothesis claims that our language influences or determines our thoughts. This view is very often regarded as an antiquated one that is backwards in construction. Indeed, prima facie, it must be the concept that came first and then the word. Language, however, is entrenched in subtleties that often escape our conscious examination of it. Using an example of a common English verb, I will try to show how much more complicated and dynamic the process is.

        Try to define the English verb "to drink". We drink coffee, milk, water, lemonade and it would seem natural to define it as an ingestion of a liquid. We do not, however, drink all liquids. Liquid medicine such as Tylenol is taken - "Have you taken your Tylenol yet?", not "Have you drunk your Tylenol yet?". Also, toxic liquids such as lighter fluid or cyanide are "Fatal if swallowed", not "Fatal if drunk". So perhaps the verb "drink" is only applied to those liquids that are naturally supposed to keep us healthy. Yet even this is not true because we can "drink tequila till we pass out" and that certainly is not healthy.

        The central facet of "to drink" must lie elsewhere. There is a distinct connotation in "drink" that is absent in "swallow". Hopefully, you have answered the title of this essay already as we will now use it help us find this facet. Historically, Socrates died by "drinking hemlock". People do not usually say that Socrates died by "swallowing hemlock". An implication of "drink" is that the person doing the drinking is doing so voluntarily. Thus, it is possible to say that someone is drinking lighter fluid, but only where the implication is that the person is doing so voluntarily and not by accident - "Crazy Uncle Bob is drinking lighter fluid again".

        Not only must the person be doing it voluntarily, he must also be doing it himself, and this is important because it is not the case in "swallow". If a chained person had his jaw forced open and had hemlock poured into his mouth, most people would say that he was "forced to swallow hemlock" and not that he was "forced to drink hemlock". If we say that someone was forced to drink something, it implies that the person was the one who held the cup in his hand and poured the liquid into his own mouth.

        So now we can define "to drink" as the act of orally ingesting a liquid, other than liquid medicine, which a person voluntarily performs himself. The liquid medication exclusion feels inelegant unless we can give a good reason why it should be excluded. And there is one.

        Notice that we "take a vaccination shot", "take Claritin and Allegra", "take liquid Tylenol or Nyquil". It appears that those medical processes that involve the addition of a chemical to our bodies uses the verb "take". So the reason why liquid medicine is excluded from drink is because "take" has jurisdiction over those kinds of medical processes. It also implies that the verb "take" is processed before the verb "drink" in our everyday sentence construction or else we would drink liquid medicine.

        So every time we use the verb to "drink", our cognitive processes must have filtered it out from a list of several other verbs and has given notice to the necessary conditions of voluntary action, liquid substance and non-medicinal substance, and have done the same with those other competing verbs. And "drink" is just one common verb. The net cast over our thought by the words we use must be present and active. It is true that our concepts must have given birth to words, but it is equally true that those words gain hidden connotations through time and must in turn limit how we think. In the end, how language evolves is much more dynamic than just a one-way street.