Conceptual Prototypes Theory




We often confuse prototypes with the "most aesthetically pleasing" exemplars. For instance, in my mind (especially if I work with a "desk linguistics" approach), I could be tempted to model the prototype of "cup" after whatever I find the most geometrically pleasing exemplar to be (with the most appealing proportions for my subjective appreciation):

It is tempting to think that the above cup IS (or at least, instantiates) the prototype, since it looks so aesthetically appealing in many senses: e.g., a) clean design, b) low cognitive load, c) "cute", d) minimalistic outline, e) curvy but not too curvy (in contrast, a mere square of 4 lines is more "boring"), etc. YET, THIS doesn't mean that when people think of cups in "abstract" (without context), THIS is what necessarily "pops up" in their/our minds (it may/might do so occasionally, but in most cases I doubt THIS is it).

Unless you are exposed to the specific design above very often in your daily life (my assumption is that most people interact daily with OTHER types of cups), chances are that your brain has formed another "picture" (another set of characteristics based on your personal experiences), which is not necessarily too far, but not too close either, to THIS one above.

I do not have any particular design in mind for now as A) this task calls for true field work (not "desk linguistics"), and B) the prototype surely differs from one generation to another, and from one family, social stratum, or geographic location to another. BUT possible candidates to be considered could well be some of the following (I offer these samples as mere improvised examples, since I have no formal field research to back it up, for now); also, bear in mind, I start this discussion with "cups," not "mugs," and that I am just guessing what types of "cups" are most commonly encountered by the "general users of cups":

In the end, the "prototype" of a concept is a complex interplay between A) what my personal experience has made me believe that "something IS"; B) my expectations of what I think other people expect "something TO BE," and C) what I think people expect from other people for "a thing TO BE."

If my wife suggests we should buy a set of cups for a friend or relative, in principle all of the above will be in my mind while I ponder whether that is a good gift choice or not. If I want a set of cups for me and my wife, a somewhat different array of characteristics could come up in my mind (which may change even further if I consider my kids as well). And if I want to buy a set of cups for myself (for my own pleasure), another set of expectations pops up.

Ultimately, what I wish to point out in this section is that, especially as educators, we often fail to realize that conceptual prototypes need not adjust to a particular aesthetic preference. I have seen too many presentations and publications in which a small white cup is presented as the best instantiation of the prototype of a cup, but I think such selection follows a personal pick of what we expect to "look nice"  rather than whatever actually lurks in the mind of the average speaker.


Quite frequently, our theory of conceptual prototypes picks out examples that are both mentally unrealistic and excessively problematic for theoretical purposes; in other words, "the conceptual prototype of our conceptual prototype" is often based on examples that deviate too much from the majority of the other prototypes. For instance, we may well discuss "dogs." What is the prototype of a dog? This proves extremely difficult to find, since there are so many breeds nowadays (10,000 years ago, this would not be such a big deal, but it is now). Take a look at all the following very "common" types of dogs:

With so many shapes of ears, snouts, and hair, as well as sizes, tempers, and colors, how could we ever pick or draw the ONE exemplar that "represents" them all?

This is an "artificial problem" (quite a common problem of desk linguistics). The human mind (in its natural habitat) does not commonly deal with such disparage diversity in the formation of a concept.

No doubt, as theorists, we delight ourselves in playing and indulging with concepts such as dogs, chairs, birds, etc. (since the current world offers so much diversity where to pick from). But regular people need not engage in such problematic situations. A classic example of many linguistics classrooms of semantics is CHAIR: Is an "ice chair" a chair? Is a "one-legged chair" a chair? Is a toy chair a chair? We proceed to ask our students: If I can't sit on it, is it a chair? We love problematizing the problem of "attributes"; we take so much pleasure in asking our students for a list of characteristics of chairs, only to come up with counter-examples for each trait.

But in daily life, we have no problem to identify a chair since they are mostly in dinning, meeting or waiting rooms (meanings do not float in the vacuum as so much desk linguistics seems to fail to realize). How we identify a) chairs in places other than dining rooms or b) weird chairs is no doubt interesting, but it does not mean that such artificial examples deny the existence of a semantic prototype of chairs (again, 10,000 years ago, this would be more evident; in the other extreme: what will it be 10,000 years from now?); such artificiality is way too often presented at the center of conceptual prototypes theories; such examples are chosen to exemplify the prototype of the prototype theory (leaving students all too often confused, with the sentiment that this is an insurmountable problem that will never be resolved).

I cannot claim yet to have an alternative when it comes to choosing the prototype of the prototype for conceptual theories of prototypes (remember this is work in progress), but I firmly believe this is an issue that should be addressed more adequately and carefully in future publications of prototype theories (much more especially so in introduction text books for students).

Without a doubt, it is interesting to analyze such extreme, peripheral examples (dogs, cats, chairs, etc.) in a theory of conceptual prototypes, but they should not be presented as the most common exemplars of the theory. Once students leave the classroom or finish reading an article text or a book chapter, they/we often leave thinking of birds, chairs and dogs as the prime examples of what "prototypes" are... Quite on the contrary, our mind is populated by a vast set of much more uniform concepts that are not internally as chaotic as the current literature would have us believe.

Ostriches, one-legged chairs and purple chihuahuas should undoubtedly be discussed, but they should not be set as the core examples of the theory; on the contrary, they should be analyzed as artificial discussions only meant to explore the limits of the human ability of categorization.

I have no field work to support the following examples, but they MIGHT serve as better prototypes of the prototype (they do show some internal variation, but they are much more uniform and could better represent the nature of what most prototypes "look like"): Hands, human skeletons, toilet paper, toothpaste, flat TVs, keyboards, wood sticks, sheets of paper for house printers, baseball head caps, pockets, jeans, shoelaces, clouds, hamburgers, spaghetti, tires, seat belts, car plates, body soaps, razors (Gillette style), compact and Blu-ray discs, flags, chimneys, planets, stars, passports, etc.

Other concepts that are tempting to be taken as prototypes of the prototype in a prototype theory, but whose nature is internally too problematic to achieve such a job are the following: guns, cars, planes, houses, trees, doors, window frames, hospitals, etc.

Concepts that I suspect are somewhere in the middle: keys, tennis shoes, hand gloves, scarves, laptops, books, eyes, marvel heroes, buttons, etc.

Final note: In the end, I think the "truth" (whatever this may be) is most probably somewhere in the middle between extremely heterogeneous concepts on the one hand and extremely homogeneous concepts on the other hand. Furthermore, an analysis in terms of reticulated semantic networks (AKA, family resemblances) is called for as well.

Prototypical? Idealized? Stereotypical? Marketing? Aesthetic? Typical? Summative? Localized? Regional? Sub-typical? Full? Complete? Unhealthy? Industrialized? Subliminal? What kind of breakfast exemplar is any of these?


As I inspect different variations in the thickness and form of the handles of cups-mugs, it is tempting to suggest that FUNCTION is the most important and defining property to consider as the essence of "cuphood" or "mughood." But trying to narrow down a concept to one ideal essential trait to define a given concept is like trying to come up with the one defining feature or trait that designates-determines-specifies what a human is (absurd, as I will try to show).

Observe the following variations of handles, starting with a few examples having to do with thickness alone:

Now, let's observe how the figure (shape) of the handle may vary in accordance to a specific "theme":

Next, let's see how some "weird" handles may appear as well:

Now, let's think of the opposite of thick handles; another possible "sub-category" I can propose is "thin handles":

And "circle handles":

And transparent mugs/cups with of course transparent handles:

Finally, "random" cups/mugs with "random handles" for which I just can't come up with a "good category" (or should I say "tag"?):

So, "thickness" alone is not the only possible variation; we can construe as many "sub-categories" as our imagination can come up with.

At some point, these "sub-categories" certainly must intersect, but what I have tried to show is that proto-typicality cannot be narrowed down to one essential feature; much less so, can we reduce proto-typicality to "aesthetic" subjective criteria as so often seems to happen (unintentionally) in textbooks dealing with conceptual prototypes theory; ultimately, it is all a matter of conventionality (and the continuum of a complex interplay between 3 mental representations: A) what my personal experience has made me believe that "something IS"; B) my expectations of what I think other people expect "something TO BE," and C) what I think people expect from other people for "a thing TO BE").


Stereotypes is another interesting case to be discussed. There is a corresponding radial conceptual prototype of "stereotypes" with nationalities-ethnicities and social groups (e.g., professions, religious groups, age groups, etc.) at the center. From there, other instances can potentially radiate outwards: In a "second level" there are stereotypes of sexual orientation groups, disabled people, and physiognomy (fat, thin, ugly, beautiful, etc.). In the periphery, as bad examples of the "prototype cline of stereotypes" (we "never" think of these as examples of "stereotypes," but we could force the concept of "stereotypes" upon them if we found a reason to do it), we have everything that is non-human: Plants, furniture, means of transportation, animals, microscopic beings, gadgets, brands (Coca-Cola, Ford), etc.

In sum, "prototype" is the more schematic concept that we have conventionalized in linguistics in order to refer to just about any concept; in contrast, "stereotypes" are a more particular type of prototype, one having to do with human-social (most often, but not always moral) judgments.

Common stereotypes of nationalities-ethnicities:

Common stereotypes of professions:

Common stereotypes of social roles: