What is language?

First and foremost, any definition of language relies upon a (conscious or unconscious) prototype of what counts as a “definition” or not; in other words, we have to ask ourselves WHAT IS THE DEFINITION OF "DEFINITION”? Is "definition" simply a matter of categorization and listing attributes? As should be evident further down below, my approach to "language" is not constrained by one single monolith of conceptualization (definitions are a human invention; reality is more complex than anything we "define" it to be); instead, I advocate for a cluster of interconnected approximations to the definition of language.

Secondly and perhaps more importantly for the matter at hand, "we the experts" tend to hold more than one representation of language, often times without us being fully aware of our own internal range. In practice, one or more prototypical definitions of language coexist in our minds with other less central conceptions of language.

With these caveats in mind, I offer seven approximations to what I think language is or may be:

1) The number one definition (central prototype) that we commonly confer to language is “means of communication,” which is heavily shaped by the conduit metaphor: “means,” “medium,” “channel,” “intermediary,” “avenue,” “route,” “path,” etc. of communication.

2) A secondary, less often discussed, yet comparably important definition that we hold in our minds is that of a “conceptualization system.” That is, a system to think about the world, a system that “structures and builds our reality”: how and what we think the world is.

3) A third conception of language is a “social construct.” In this definition, language is what it is only relative to its role in the formation and maintenance (sometimes destruction) of social bonds. Language is “whatever is shared by the members of a society”: common ground, mutual understanding, meaning negotiation, etc.

4) A fourth meaning associated with language is “set of linguistic symbols.” This places an emphasis on the symbolic nature of language. Different definitions of “symbolism”/“symbolization” come into play though. Some would go as far as to tell that human beings are “symbolic creatures,” which is not untrue, but such definitions lean to a rather ethereal shade of conversation. Others are more technical and define symbols as arbitrary units of mental representation: Fully detached “signs” for the representation of referential and non-referential objects of conception.

5) A fifth idea linked to language is “words.” This is probably the, ironically, least self-conscious definition of language. Hardly anybody would deny that language has words; yet, not many would be willing to give such a seemingly barbaric definition; however, in practice, our daily experience with language is an experience with words (which is what you are reading right now (funnily enough)). So, it is just fair that we have this unaware conception of “language as words.”

6) Quite inadvertently as well, our conception of language heavily depends on the ideas we indoctrinate ourselves with from our favorite authors. The more I read and the more I agree with a certain author, the more I will tend to define language in her or his own terms.

As can be seen thus far, language is liable to different conceptions and definitions. Different metaphors can be evoked to try to organize these different ideas into one coherent prototype cline. The one that seems most comfortable and practical (both conceptually and didactically) to me is the “radial metaphor,” in which we have a central core definition of language (the prototype); and then from there, other definitions extend outwards towards the periphery:

An alternative metaphor that I like to use in order to think of definitions is the “reticular meaning” metaphor (Gontier, 2016). In this metaphor, meanings overlap to a far more reaching extent than we normally are aware of. The following picture is a simplistic illustration of what the term “reticulate” is intended to refer to.

On the left side, different meanings of a prototype branch out separately; on the right side, meanings interact and interweave with one another as they unfold along various interconnected ramifications.

A third alternative metaphor is the “waves metaphor”:

In this metaphor, different aspects of meaning can peak or valley at different times in our brains.

And yet another metaphor, perhaps the most insightful and elaborate of all (with the most explanatory power), is that of family resemblances, as suggested by Ludwig Wittgenstein (1953). The "Smith brothers" example (by Armstrong et al., 1983) should suffice to illustrate this. As can be seen below, each member of this family possesses traits that are also manifest in other members of the family; yet, no single member bears all traits at once; what is more, no trait is shared by everyone. In any case, the man in the middle brings forth the highest number of shared features, so he could very well be picked as the most representative member of this category.

The Smith Brothers.

Original drawing in Sharon et al. (1983).

This drawing by Reisberg (2022).

In this fashion, the prototype of this particular category (not necessarily true for all categories) is an abstraction of all the members, but the prototype per se does not exist (is not materialized) in one single member. In a similar vein, perhaps language is never realized or materialized as everything that it may be all at once. Perhaps different facets of it emerge at different times in different situations.

So, what is language, then? I do not actually intend to deal with phenomenological or ontological issues. This is just a Wittggensteinian approach (the meaning of words is its conventional use). But I do want to point out two facts: 1) Those of us who work as language teachers are less likely to succeed and are much more ineffective if we do not ask ourselves what language is in the first place; and 2) our conception of language is not one single monolith.

7) Personally, my favorite definition is that “language is the evolutionary mechanism that humans developed to achieve ‘shared cognition’ (a common consciousness with the others).” But that is something I would like to discuss at some other time, since this approach calls for much more fine-grained distinctions from the cognitive neurosciences, such as sensation, perception, and conception; this text hereby is only intended to serve as a brief reflection upon the conceptual range of definitions of language that currently prevail in our field of work.


Language as a means of communication: Reddy, M. J. (1979). The conduit metaphor: A case of frame conflict in our language about language. In A. Ortony (Ed.), Metaphor and Thought (pp. 284–310). Cambridge University Press;

Language as a conceptualization system: Talmy, L. (2000). Toward a cognitive semantics. MIT Press;

Language as a social construct: Silverstein, M. (2004). “Cultural” concepts and the language-culture nexus. Current anthropology, 45(5), 621-652.

Language as symbolization: Langacker, R. W. (2008). Cognitive grammar: A basic introduction. Oxford University Press.

Language as words: Kenneally, C. (2007). The first word: The search for the origins of language. Penguin.

Language as an indoctrination: Harris, R. A. (2021). The linguistics wars: Chomsky, Lakoff, and the battle over deep structure. Oxford University Press.

Language as shared cognition: Ivry, Richard (2009). Lecture 24. Cognitive control: Working memory [Video].  In Cognitive Science C127. UC Berkeley.

Final note:

Excerpt from Langacker (1987):


Armstrong, S. L., Gleitman, L. R., & Gleitman, H. (1983). What some concepts might not be. Cognition, 13(3), 263-308.

Gontier, N. (2016). Guest-editorial introduction: Converging evolutionary patterns in life and culture. Evolutionary Biology, 43(4), 427-445.

Langacker, R. W. (1987). Foundations of cognitive grammar: Theoretical prerequisites (Vol. 1). Stanford University Press.

Reisberg, D. (2022). Cognition: Exploring the science of the mind. WW Norton & Company Incorporated.

Tuggy, D. (1993). Ambiguity, polysemy, and vagueness. Cognitive Linguistics 4(3): 273–290.

Wittgenstein, L. (1953). Philosophical investigations. Philosophische Untersuchungen. Macmillan.