If humans could create any kind of smell for someone else to experience. Could these smells be used to form a whole language?

Language creates words by different sounds, so would it be as simple as chaining different sequences of smells in time to form different words?

Like maybe: pumpkiny+fishy forms one word and pumpkiny+fishy+rose forms a different type of word.

Or maybe each smell is it's own word on its own? I'm not sure what the limit of distinguishable smells are or if how the information contained in smells can be efficiently broken up into distinct information blocks for language.

  • 5
    This kind of question is better put on Constructed Languages – jk - Reinstate Monica Jun 23 at 10:09
  • People's senses of smell are quite variable. There are, for instance, many sorts of smell-blindness, analogous to color-blindness, How could a "hearer" of some sentence of smell-language figure out what a "speaker" intended if the smell senses of the two are incommensurate? – Greg Lee Jun 27 at 2:59

It seems unlikely that you could use smell in a similar way to voice or signing. Natural languages tend to be transitory - or sounds and sights do not hang around in the air that long, allowing new segments of the language to be prominent. Smells don't disperse quickly, so instead of having a linear sequence, you'd have a mix of old and new smells with no real distinction. Even in written language, there is a sense of linearity that uses the reader's focus to only see one part of it, simulating that transitoriness.

| improve this answer | |

Obviously this is extremely speculative, given the premise, but yes

At a trivial level, with 26 distinct smells, you can implement English over scent by having one scent for each letter of the English spelling

Language can be communicated through almost any medium which can distinguish at least two states (this very post goes through such a medium, albeit not in a very human-friendly way, on its way to you) and a language natively used in a particular medium will behave "phonologically" in a different way due to the features of that medium. Unfortunately, we only really have two test cases for this in the real world: spoken language (through the medium of sound and with a fairly small pool of available symbols), and sign language (through a visual medium with a very large pool of symbols)

Writing does not count as there is no language that is natively written but not natively spoken or signed, but it may still be a useful point of comparison. Like sign languages, it's a visual medium with a very large pool of symbols so a purely written language would probably be expected to behave "phonologically" more like a sign language than an oral one. Compare manually-encoded English which is largely analogous to logographic writing (i.e. signs represent words), but with some alphabetic supplements (i.e. finger spelling)

A scent-based language, could behave more like an oral language, or more like a sign language, probably depending largely on the number of available scents

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.