Donkeys Uncanned

A Tale Of Dictionaries And Reactionaries

Kim Hipwell
11 min readJan 15, 2022

What This Is About

I’m going to examine how language behaves outside of the confines of dictionaries, and show how that behaviour undermines reactionary rhetoric about word meanings (with a specific focus on transphobia).

Paraphrase As Definition

Dictionaries — compilations of circumlocutory paraphrases — do not authoritatively tell us all there is to know about what words mean.

They cannot, because word meanings change over time.

What dictionaries do, with varying levels of success, is to capture cliches: their “definitions” simply illustrate well-evidenced historical patterns of word usage. But constant updates are required to maintain a dictionary’s currency as those patterns shift: words and senses are added to reflect the emergence of new meanings, obsolescent ones are discarded.

Dictionaries only reflect how words have been used; they do not arbitrate on how words can be used.

Generative Meaning

The Speed At Which Meaning Shifts

Although it’s obvious that language changes, those changes are often viewed as being the result of a kind of evolutionary process, emerging from gradual drift over long timescales.

In this view, language shifts course like a river, slowly meandering through ambiguities; new senses of a word emerge as incremental shifts in word usage accumulate, requiring dictionary definitions to be tweaked from time to time. (Which allows those definitions to be seen as persistently authoritative, if sometimes perhaps slightly fusty).

But this is an erroneous view of how new meanings emerge; or a partial one, at least. There are many examples in which new meanings for a word, ones that are radically disconnected from any existing pattern of use, are generated on-the-fly. Language moves more like a landslide in these cases, shifting near-instantaneously.

A Just-So Story

I’m going to use a just-so story to illustrate how these generative meanings emerge: a fictional example helps to distance us from language usage we are already familiar with, but we’ll double back later to look at some real world examples.

Here’s the background:

A donkey escapes from a field, and causes havoc on a nearby motorway, leading to the biggest traffic jam in history, a weekend of chaotic nightmare for all concerned. This becomes a huge news story; the donkey that caused it all becomes famous, an emblem of chaos.

Emblems of chaos

As a result, people start to use the word “donkey” to refer to obscured causes of substantial delay (“not sure why the computer is taking so long… must be a donkey in the works”). Common knowledge of the legendary traffic jam allows for the import of “donkey” in such cases to be pragmatically inferred.

From this point, new senses can proliferate though mechanisms such as metonymy or analogy, with one meaning being chained together to the next. Maybe “donkey” starts being used to label someone who is consistently late, holding up others; from there, for anything that misses a deadline; for sports that run into extra time; for anything that feels slow; and so on.

In this way, an entire new set of meanings can spontaneously arise for “donkey”, radiating from the central new use — all the result of linguistic creativity, all dependent on a single “legend” event.

{ Sidetrack: in this example, our “legend” is a news story, but any kind of widely shared narrative would serve: a scientific discovery, urban legend, joke, meme, song lyric, movie, TV show, video game, or other glitch in the matrix. }

Generative, Not Figurative

Note that our new patterns of use for “donkey” haven’t arisen through analogy to previously stereotyped characteristics of donkeys.

In other words, these new senses are not the result of “donkey” being used figuratively to indicate a stubborn or stupid person, someone displaying donkey-like intractability (which is an analogy so commonly used that the adjective “asinine” — meaning “pertaining to donkeys” — rarely refers literally to donkeys).

Some of the proliferation of new senses may be the result of figurative connections, but they are connections with the newly generated meaning, not stemming from previously established uses.

So, sense extension of this kind does not rely on anything intrinsic to the pre-existing meaning(s) of a word: it can arise due to a one-off happening, an entirely contingent event. A slight change to the originating “legend” would have allowed for “monkey” or “turkey” or “custard” — any word for something that could have caused traffic jam chaos — to accrete those new senses.

Examples Of Eponyms

Although the “donkey” example is fictional, this kind of generative extension of word meaning is commonplace in real language.

For example, there are a set of words called eponyms, which emerge when proper names are taken as pointers to a “legend” which motivates the generation of new meanings for those words. Here are a few:

  • sandwich — the foodstuff legendarily named after the snacking preferences of John Montagu, 4th Earl of Sandwich. Via sense extension, this provides a generic word for layered constructions of all kinds.
  • gaslighting — forms of psychological abuse, derived from Patrick Hamilton’s 1938 play “Gaslight” (and/or later films of the same name).
  • catfish — to masquerade on social media in order to target a person, taken from the film of the same name (the name of which is ultimately derived from a “legend” about cod and catfish being shipped together).
  • scrooge — a mean/miserly person, particularly one who hates Christmas, named after Dickens’ character (the use of an archetypal name to label a person with a character trait is specifically called anotonomasia).
  • stan — an obsessive fan, also used as a verb for experiencing such obsession; a word originating from the Eminem song “Stan”, in which the protagonist’s fury at being “ignored” by his hero leads to tragedy

This process can even give rise to morphological particles, such as -gate (deriving from Watergate), frequently used by journalists to indicate a scandalous, career-ending revelation.

Eponyms emerge with great regularity. Product names are susceptible to this process: businesses actively guard against the “erosion” of their trademarks, aiming to prevent genericization (a name becoming the generic word for a product category).

The word “escalator” is a product name, genericized

Pervasive Generativity

There are many other examples of creative word use in language that show us that word meaning is a highly renegotiable, slippery thing, often subject to improvisations in the instant.

Consider, for example, repurposing of words across grammatical categories: in English, “verbing nouns” and similar shifts are commonplace. Or the rapid evolution of meaning within “multiethnolects” and new sign languages. Or the creative use of words within metaphor and humour.

Fundamentally, semantic compositionality seems to rely on context being taken into account; the meaning of a word is not simply encapsulable.

Language is not a mere stringing together of a fixed repertoire of canned meanings (asino morto): both expression and interpretation are flexible, creative processes in which an understanding of shared societal references play a crucial part.

“But Then Nothing Means Anything”

Words And Motivations

The generation of new meanings, although not at all predictable, must be well-motivated. There has to be a shared history, an analogy that can be made, or an existing pattern of use that binds how a word is used: obviously we can’t use simply any word to refer to anything.

The fact that language moves, whether that be as a river or as a landslide, means that closely examining those motivations helps us understand not only how and why we conceptualize the world and self, but also how those conceptions are subject to change and challenge.

But there is a reactionary strain of thought that holds that such analysis is suspect, that (somehow) it implies that words are effectively meaningless. Consider this example from Kathleen Stock, prompted by Judith Butler examining the motivations for the ways in which we use the words “gender” and “sex”:

[From “Material Girls” by Kathleen Stock]

This treats as if the words we use in modern English correspond neatly to the ontology of the universe; as though the author believes that their current patterns of use provide a perfect mirror of the structure of reality, and that any interrogation of those patterns is an attack on that entire structure.

This is breathtakingly gauche, a hick’s view of word meaning, unworthy of oxygen — or even phlogiston. The naturalist Stephen Jay Gould, in pellucid prose intended for the layperson, spells out the issue from a scientist’s perspective:

[From “Full House” by Stephen Jay Gould]

It makes no difference how “scientific” the vocabulary that we consider is. Scientific progress is tied to conceptual revolutions; its terminology does not asymptotically approach a state of perfection, and it does not pick out “natural kinds” whose existence and delineation is indisputable. Science depends on its conceptual infrastructure being revisable, and therefore disputable.

But Stock is attempting to cement a (transphobic) worldview, not to grapple with dissonances between world and words. In service of cosy folksonomic certainties, she wishes to insist certain “biological” words can be eternally conjoined to an authoritative dictionary definition. But that idea can’t fly.

Particularly in the field of biology, the notion of “natural kinds” is a very hard sell indeed. Evolutionary theory is based on the mutability of kinds, and developmental biology shows us a variety of the mechanisms that support plasticity of development. And indeed, current science shows us that the biology of sex determination is very complex indeed, making mock of Stock’s homespun wisdom about “two naturally pre-given, stable biological sexes”.

What Are “Biological” Words Anyway?

Even if we were to trashcan objections such as Gould’s alongside the actual biological science, Stock’s schtick is bound to fail miserably, simply because “biology” is absolutely not the only motivation that determines the usage of the words she wants to contest.

Putting aside “sex” for a moment, a good parallel is that we can’t pin motivation for the uses of “mother” down to a simple matter of biological relationship. Consider, for example, that we use the word to include, inter alia: stepmothers, adoptive mothers, foster mothers, surrogate mothers, birth mothers, and biological mothers. The scope of reference for “mother” is not a matter for science — it is a matter for society.

(And, of course, remember that “legend” can instantaneously affect how “mother” and “father” are used, examples being the impact of IVF and awareness of trans fathers giving birth — as per our just-so story above, single events can have a huge impact on the way language is used).

In exactly analogous manner, a person’s “sex” does not simply equate to a biological state, such as “reproductive role” (many people don’t have one) or genome or anatomy (“sex” existed before DNA or even chromosones were known of; people in possession of a great diversity of anatomies have existed uncontroversially as “man” or “woman”).

Just as “mother” encompasses a set of well-motivated uses that connect to the sense in which “nature’s factuality offers hints”, so do “man” and “woman”.

When we are talking about “sexes”, we are therefore very often talking about social roles: the words “sex” and “gender” are used interchangeably in most circumstances, because the motivations for their uses are largely commingled.

Definition and Circumlocution

When we consider abstract concepts — promises, verdicts, twerking — we can easily see that they are not “impositions” on a pre-existing natural world. They are part of our social world; these are entities actually brought into being through our use of language or other signs.

(We can see that this also applies to the eponyms we listed earlier; even a concrete, tangible and tasty treat such as a “sandwich” is an abstraction in these terms, a member of a created category, not an “imposition” on a pre-existant wild smorgasbord).

The science of sandwichology

As we’ve already noted, “sex” encompasses social role rather than simply being a biological status, meaning that sexes necessarily must be similarly abstractions, constructed in just the same way as “mother” is. This is rather less intuitive for many, and is the idea that Stock is aiming at when she is ineffectually trying to ridicule Butler. (Seen in juxtaposition with “mother” being an abstraction, the idea does not seem caustic enough to dissolve the entirety of science).

Dictionaries do seem to offer a comfortingly concrete riposte to such ideas.

However, attempts to impose reactionary ideas by citing a dictionary definition are doomed to vacuity, simply because a definition is nothing more than a circumlocution, a restatement of what the word means.

For example, take the definition “adult human female” (AHF) that is used to argue about the inclusivity of the word “woman”.

There is nothing “wrong” with using the AHF definition, just as there is nothing inherently wrong with using the numbers “14” or “88”. But reciting “adult human female” has become used simply to indicate the speaker’s prejudice against trans people, in just the same way as those numbers are used to indicate allegiance to Nazi ideals.

But using AHF as an “argument” is vacuous because two people who both accept the AHF definition can have very different views about who is included in the scope of “woman”, which will be straightforwardly be mirrored in their assumptions of a matching scope of reference for AHF.

Trans hostile people seem to assume that “female” solely specifies something “biological” that would vindicate their vindictive viewpoints; it simply ain’t so. And even the dictionary (Mirriam Webster) will back me up on this:

In other words, the AHF definition of “woman” does nothing to arbitrate on boundaries of reference — it doesn’t limit who the word correctly describes. Even more fundamentally, an existing definition can never limit how a word should be used; society decides how the word can be applied.

Those naïfs who believe they can nullify societal acceptance of trans people simply by reiterating the self-referential loops of a dictionary on a t-shirt or a billboard exemplify what it means to be “inane”.

In A Nutshell

The generation of entirely new and unpredictable meanings on the basis of “legend” demonstrates the incessant creativity and playfulness that propels language: the antithesis of the stasis that dictionary definitions imply.

This shows us that reactionary views of what “meaning” is are hopelessly naive, based on assumptions about an imaginary stability of word use and authority of definition that doesn’t cash out in the real world we live by, the social world.

Meaning is something we create, not something we excavate.

--

--

Kim Hipwell

PhD in Cognitive Science, interested in the structures of natural and artificial languages. Thrives on atonal music and trans rights. She/her.