Friedrich Nietzsche is well-known for his denunciation of objective truth, and even more well-known for his cynicism. His views on language are certainly not spared from this cynicism; his comparisons of language to columbariums and cobweb cathedrals reek of death (even as he uses language to make those comparisons). Regardless of whether one chooses to interpret this as irony or mean-spiritedness, his arguments on the nature of language reveal much about its social and experiential implications. The variety of human experience largely confirms his views, even while complicating the story he tells. 

According to Nietzsche, language is essentially a structure of “tokens of designation” used to represent our experience of reality.¹ Reality, in his view, is chaotic and fluctuating with no underlying order between individual phenomena. Language is humanity’s way of “digesting” reality by artificially imposing order on it.² The units of language, the individual tokens of designation, are words, which he calls “a copy of a nervous stimulation in sounds.” Words themselves act as metaphors by equating the sounds our mouths form to the non-physical concepts we impose on reality._³ Through the act of categorizing our experience of reality under concepts and regurgitating those concepts as sounds, those experiences lose their individual particularity. As Nietzsche puts it, we “dissolve an image into a concept.”⁴ In assigning concepts to experiences, we categorize individual events of reality that are fundamentally incomparable under the same label.⁵ No two events in the physical world are ever the same; they each happen in different times and places, with chance variations that differentiate them. The act of categorizing these experiences entails determining which differences mark boundaries of concepts, and which can be overlooked for the sake of similarities. There is no objective way to determine which differences are negligible, since all differences between things are objective in themselves. Therefore, the differences that determine conceptual boundaries become a matter of social consensus. By this model, language is a socially constructed “army of metaphors.”⁶ Correct representations of reality are defined by an “obligation to use the customary metaphors” for experiences that the social construct prescribes.⁷

Nietzsche points out that because of this, language depends on a potentially disprovable assumption: that people have similar enough perceptions of the same reality that their experiences can fit within this socially constructed conceptual framework. Communication requires that others have the ability to conceptualize reality, and that the reality they perceive is similar to other’s perceptions.⁸ People need to have the same basic experience “behind” a concept so they can have a shared understanding of the concept’s meaning. In this view, the stability of language and, by extension, the world we conceptualize is dependent on shared perceptual modes. The example Nietzsche uses is that of science, where this is most obvious.⁹ Scientific understanding would be impossible if observers did not have the same perception of the same objects. In studying something scientifically, one is forced to assume that their basic mode of perception is objective. Otherwise, scientific findings would only be applicable to an individual observer’s perception of reality, rather than being relevant to human experience. Nietzsche exposes the flaws of this fundamental assumption even as he describes it: 

…If each of us still had a different kind of sensuous perception, if we ourselves could only perceive things as, variously, a bird, a worm, or a plant does, or if one of us were to see a stimulus as red, a second person were to see the same stimulus as blue, while a third were even to hear it as a sound, nobody would ever speak of nature as something conforming to laws; father they would take it to be nothing other than a highly subjective formation.¹⁰

The stability behind language, or any conceptual mode of organizing reality, is easily disproven if people don’t have common modes of perception. The driving assumption of language becomes disturbingly flimsy; the social currency of words has no consistent backing if this is the case.

There is one easily accessible example that serves as a counterexample to this stability, one that isn’t removed from humanity like the birds and worms Nietzsche references: autistic perception. Autistic perception is often wildly different from that of the non-autistic (or “neurotypical”) majority. On top of this, autistic perception can vary widely even among autistic people themselves. One neurological study showed that while neurotypical neurology is structured relatively consistently, autistic neurology differed widely in how different regions of the brain interacted. When functional interhemispheric connectivity was studied in autistic participants against a neurotypical control group, the authors found that individual participants had both hyperconnectivity and hypoconnectivity in varying regions of the brain compared to the control.¹¹ In other words, autistic brain connectivity varies widely in its “idiosyncrasy” between autistic participants, while neurotypical brain connectivity is relatively consistent._¹² So while it may be relatively accurate to assume common perception in the case of the neurotypical majority, autistic subjects do not share common perceptions with each other or with the majority.

Because of this phenomenon, autistic people are often very Nietzsche-ian in their understanding of concepts and experience. They are well aware of the fluidity of experience, and the fact that events in reality are constantly novel and incomparable, because that is reflected in their sensory perceptions. The socially-determined framework of relevant versus irrelevant distinctions does not work for autistic people; all differences are equally perceived, or the ones they’re inclined to focus on do not match the neurotypical method of conceptualization. Ralph Savarese, a teacher of autistic literature students, describes this succinctly: “To an autistic child, the wind in the birch trees is just as compelling as a friend’s comment on the playground.”¹³

These examples of autistic perception serve as a case study for Nietzsche’s refutation of perceptual stability. As he says, “If each of us still had a different kind of sensuous perception… nobody would ever speak of nature conforming to laws.”¹⁴ The large variety in modes of perception throws into doubt any idea of “nature conforming to laws;” it’s clear that each subject’s perception is vastly different, with few commonalities that would qualify as an objective standard. If the idea of common perception that language is based on is so doubtful, then the stability of language is thrown into chaos. This is reflected in autism’s characterization as a social-communication disorder, rather than a strictly sensory one. When autistic people do not share in the neurotypical majority’s idea of common perception, their perceptions often cannot be translated into the language based on that common perception. The currency of language has no experiential backing in their case; the concepts that are meant to contain one common experience will have no meaning to someone without that experience or may evoke an entirely different experience. Understanding quickly breaks down. 

While Nietzsche’s argument on the nature of language holds merit, it is ultimately founded on something that he does not articulate within his essay: the complete separation between thought and perception. In Nietzsche’s model of language, thought (which he equates with the formation of concepts and language) is imposed on perception in order to “tame” it. This view requires that thought is something foreign to perception. The interaction of thought with perception is very linear in his view: raw perceptions of reality are “digested” and sorted into categories of thought, which then take the form of language in words. Thought, in the form of language, is used as a metaphor for reality as we perceive it, but there is no natural overlap between these two categories. As he describes it, there is a “complete leap from one sphere into the heart of another, new sphere” when thought is imposed on perception.¹⁵

This assumption on thought and perception echoes Immanuel Kant’s views on the subject in The Critique of Judgement nearly a hundred years earlier. Kant articulates the distinction more clearly, though he uses the terms imagination and understanding instead of perception and thought. Imagination is the capacity of perception, which deals in images formed from raw sensory experience. Understanding, on the other hand, is the capacity of thought and traffics in the concepts and categorization that Nietzsche describes. Both are necessary for cognition, but according to Kant, they are completely separate faculties that cannot overlap. In his view, they can only be brought into conversation with each other. The faculty of judgement can reconcile them, but successful judgement is marked by a harmony between the two fundamentally separate functions, rather than their combination. Nietzsche assumes a similar view of thought and perception, though he doubts any straightforward possibility for their harmony. He believes that any application of thought to perception marks a departure from reality. In his words, thought is like building “cathedrals of concepts on… flowing water.”¹⁶ Because the nature of reality and perception is so fluid, it is inherently iconoclastic. Concepts need to be removed from reality in order to have any stability. In this case, harmony would be impossible. But regardless of Kant and Nietzsche’s views on whether thought and perception can be reconciled in the end, they both depend on the idea that the functions are terribly separate.

While the example of autistic perception reflected Nietzsche’s view of language as a social construction, it deeply complicates the core premise that Nietzsche’s argument assumes: the complete separation of thought and perception. Nietzsche associates thought exclusively with the act of conceptualizing within the framework of language. Among the non-autistic (or “neurotypical”) majority, this view of thought is relatively accurate. Concepts take the form of categories attached to words, which are developed to mediate sensory perception. The relationships between these concepts are confined to the structures and determined verbally by the framework of language. The complication is that the framework of language is not the only one where thought can occur. Within the autistic population, thought often occurs independently from language, and in many cases overlaps with perception. Thought is not merely a conceptual framework imposed on perception to make sense of it; instead, thought is organically and inextricably attached to perception. 

Arguably, this is the case for everyone, autistic or neurotypical, though on a much subtler level. The brain automatically organizes raw sensory data from sense organs, even before the perception reaches consciousness. For example, in the case of sight, the brain automatically corrects for the distortion of light waves caused by the lens of the eye before they hit the retina; without this correction, visual perception would appear “upside-down.” In this case, the “interpretation” of raw sensory data occurs before the subject even becomes aware of the perception. Another example is “sensory gating,” or the filtering of irrelevant sensory data before it reaches consciousness. This function is largely determined by arousal in the Automatic Nervous System, which is the subconscious function of the brain. The more aroused a subject is, such as in fight-or-flight mode, the more sensory data reaches conscious awareness and processing. These kinds of interference with raw perception could be described as thought, though they are automatic and unconscious. They are inextricable with perception by the physical nature of sensory processing within the brain. However, thought in the sense of conscious conceptualization can also function in a way that echoes these examples.

In her book Autism and the Edges of the Known World, Olga Bogdashina lays out a theory as to how this can work. She points out that the language, which Nietzsche associates with thought, uses words as symbols, or placeholders, in forming structures of concepts.¹⁷ Similar to Nietzsche, she notes that these symbols have no direct connection to the things they refer to. Autistic people, on the other hand, tend to deal with non-verbal indexes when forming concepts. An index is a signifier that has a direct causal relationship with the thing it is signifying. For example, smoke is an index for fire; fire does not exist without the presence of smoke, so smoke naturally indicates fire. The indexes used for forming and representing concepts have a causal basis in perceptions. Bogdashina calls these methods of thought “non-verbal languages:” perceptual frameworks for thought that are not independent of sensory experience.¹⁸ While the language framework of thought operates independently of sensory experience and is later “applied” to it, the perceptual framework operates with indexes that cannot be separated from sensory experience. She says,

Autistic children, like non-autistic ones, learn through interaction with the world, but their reality is very different. They learn their language(s) through interaction with objects and people on the sensory level. That is why their “words” have nothing to do with the conventional names for things and events that we use to describe the function of these things and events. Their “words” are literal, they store sensations produced by objects through interaction, and they “name” them accordingly.¹⁹

To illustrate the process, she uses autistic author Donna Williams as a case study. Williams describes the use of perceptual frameworks through the example of sound and kinesthetic “languages”:

Going back to the door before it is known in terms of function… it may have no word at all. Later, as one moves from non-physical sensing of what happens to be a door to a physical-based sensing of what happens to be a door, its sound-concept is very unlikely to be “door.” If you tapped the door, it may (depending on the door) tell you its name is ‘took.’ If it made a noise when it gave way under impact, it might say its name is “rerr” if it drags on the carpet or “ii-er” depending on the sound of the hinges.²⁰

She explains that thought within this framework occurs both in terms of causal sensory relationships between “sound-concepts,” and through something that resembles free association:

It may have no sound concept at all or sound concept relating to the experience of door might come from the emotional experience of sensory buzz associated with that door. So taking the example of the swinging door fascination, if the buzz experience brought out a little suppressed squeal… the sound concept associated with the experience of door may well become this stored and later triggered sound.²¹

Autistic author Temple Grandin describes the mechanics of how concepts are used in this mode of thinking, as well as it can be described in words. She points out that while thought within a language framework tends to work sequentially from one concept to another, thought within a sensory framework relates concepts by their associative sensory relationships:

An autistic man who composes music told me that he makes “sound pictures” using small pieces of other music to create new compositions. A computer programmer with autism told me that he sees the general pattern of the program tree. After he visualizes the skeleton for the program, he simply writes the code for each branch…. My thinking pattern always starts with specifics and works towards generalizations in an associational and nonsequential way. As if I were attempting to figure out what the picture on a jigsaw puzzle is when only one third of the puzzle is completed, I am able to fill in the missing pieces by scanning my video library.²²

What distinguishes these frameworks of thought from language’s framework is that concepts are constructed directly in sensory perceptions, rather than the extended “metaphors” that are language. While the language framework does operate separately from sensory experience, as Nietzsche argues, thought is not confined to the language framework. In a perceptual framework, thought is not imposed on sensory perceptions in the sense that it is with language; instead, it functions within the parameters of sensory perception. The framework of language is independent of the “logic” of sensory perception, and perception is interpreted in a way that functions within language concepts. On the other hand, when operating in a sensory framework, concepts are directly related to the sensory perceptions that form them, and thought functions directly through the relationships between sensory perceptions. 

The fact that autistic people tend to use sensory frameworks for thinking, rather than a language framework, has intriguing implications when they do interact with language as such. Many autistic people do learn to use language as one of their means of cognition, even if it is not their main or most natural one. This can be the case for both speaking autistic people and those who use methods like writing or AAC. In this case, the issue of translatability between modes of thought comes up. This issue goes back to the problem of shared perception as a basis for language and communication; shared communication requires a shared framework of how perceptions are constructed into concepts. When these frameworks are completely different, as is the case with language frameworks versus sensory frameworks, then direct translation between the two is not possible.²³ 

Temple Grandin is one of the most famous examples of autistic people who discuss this issue. Her primary sensory framework is visual; as she puts it, she “thinks in pictures.” She describes the translation process in the first paragraph of her now-famous book Thinking in Pictures: “Words are like a second language to me. I translate both spoken and written words into full-color movies, complete with sound, which run like a VCR tape in my head. When somebody speaks to me, his words are instantly translated into pictures. Language-based thinkers often find this phenomenon difficult to understand…”²⁴ Essentially, she attaches words, the units of language, to images, the units of her sensory framework of thought, in a one-to-one relationship. This is why she strongly prefers literal, concrete language; it requires less navigation in the language framework of thought before she can “convert” them to her units of thought. The fact that she is converting back and forth between a language framework and a visual framework that do not follow the same track means that her use of language can be confusing to those interpreting it within the framework of language. She uses the words “French toast” as a hypothetical example as to how this might work for an autistic child. The hypothetical child attaches the words “French toast” to the image of French toast, which evokes the experience of eating toast, which makes them happy. So they use the words “French toast” to represent happiness.²⁵ The logic makes perfect sense within an associative sensory framework of thought, but within the socially defined context of language, where the meaning of “French toast” is confined to its agreed-upon usage within a situation, the line of thought is confusing at best.

It is also possible for autistic people to interact with language through their sensory frameworks, without ever using a language framework to construct meaning from it. In this case, they interact with words as physical realities in themselves, rather than as symbols for linguistic concepts. While Nietzsche acknowledges that language takes the form of a physical reality (sound waves in the air, or ink on a surface), he does not explore the implications of this fact. Many other philosophers have written about meaning in relation to language as a physical phenomenon. For example, Ermanno Bencivenga explores the issue in a chapter of his book My Kantian Ways, which he titles “The Thickness of Words.” Though he still considers the issue in relation to the framework of language, he observes that the physical experience of words does not naturally fit into a linear conceptual system. In fact, the physicality of words often hinders thought as it occurs within this framework. He compares its physicality to a fly that buzzes around a man who is deep in thought and nearly detached from the physical world; when a person is aware of its physicality, it breaks the smooth operation of abstract thought by virtue of its connection to the physical world.²⁶ Language matters insofar as it is transparent to the meanings attached to it, and the words attached to concepts are completely arbitrary. As Bencivenga puts it, “Language has meaning, words refer to things, and that’s what language and words are good for. ‘Table’ means table, ‘chair’ means chair…. And it doesn’t matter that ‘table’ sounds, or reads, like ‘table’: we could just as well have decided to use the word ‘rable’ instead, and then ‘rable’ would mean table.”²⁷ This echoes Nietzsche’s view that words are physical sounds with no direct connection to the language framework, and the only meaning they have is the ones imposed on them through social construction. 

But Bencivenga goes further and argues that imposing the language framework on words is an act of violence. He says that in order to operate smoothly within the framework of thought, words have to be “beaten to a pulp” until they become transparent. He uses historical timekeeping methods as an example of meaning being forced onto physical phenomena used for communication:

When there were no clocks, or watches, or any other mechanical concoctions to count time, people found the most creative ways to get the job done. Some were truly sadistic. Slaves were… chained motionless, and a torch was lit up on their head, and when the fire reached the slave he would scream. And his scream would stand for a time interval.²⁹ 

In this description, he observes that the physical scream (and words by extension) has an inclination to meaning in itself and resists the arbitrary imposition of meaning from an external language framework. He describes this associative, visceral sense of “native” meaning in a way that would fit into a sensory framework of thought: “Now what does it take for a scream to work like that? A scream is full of anguish, of pain, of fear; a scream makes our blood boil, our legs run; a scream makes us want to help.”³⁰ But it is used as a placemarker in a conceptual framework and subjugated to meaning through a socially defined system independent of the thing itself.³¹ He says, “For a scream to be converted into a signpost, all other context must be annihilated…. So that only the time interval is left; only the time interval matters.”³² Words as physical sensory experiences are native to a sensory framework of understanding, but they have been commandeered by the framework of language in a process that is, like Nietzsche said, arbitrary and unnatural.

Many autistic people are well aware of this phenomenon, regardless of whether they would describe it in these terms. While the physicality of words may be a hindrance within a language framework as neurotypical people use it, autistic people tend to be much more attuned to words as physical objects. Many operate with words on a purely physical level within their sensory frameworks of thought; they experience words as “thick,” as Bencivega would describe it, rather than transparent to an imposed linguistic meaning. A large number of autistic autobiographies describe some variation of this experience, which appears to be common. Autistic author Donna Williams describes it in the context of learning to use language; she talks about language as patterns of sounds that she was perfectly content to interact with outside of the context of language. She says, “Words were no problem, but other people’s expectations for me to respond to them were. This would have required my understanding what was said, but I was too happy losing myself to want to be dragged back by something as two-dimensional as meaning.”³³

Autistic author Tito Mukhopadhyay describes this experience in more depth, as well as how he processes them within his sensory framework. Mukhopadhyay’s primary framework is sound, but because he’s synesthetic, it overlaps strongly with the visual framework as well. In a story he tells in his book How Can I Talk If My Lips Don’t Move?, an autism researcher studying language competence instructs him to listen as a book is read aloud: 

Claude [the researcher] read…. I saw the voice transform into long apple green and yellow strings, searching under the table for who knows what? Threads like raw silk forming from Claude’s voice.

Claude read. I watched those strings vibrate with different amplitudes as Claude tried to impress the silent beholders and serious researchers of autism with the varying tones of a near-to-perfection performance.

Claude read. I watched those strings with stresses and strains, reaching their own elastic limits and snapping every now and then, when his voice reached a certain pitch. I saw those snapped strings form knots like entangled silk, the color of apple green and yellow.³⁴

Mukhopadhyay processes the words through their tones and volume which also takes the form of colors and textures. The meaning he constructs from these sensory concepts involve both the words as they function within the physical environment, and the emotional context of the speaker. His understanding of the words reflects Bencivenga’s description of language well. It is a very obvious example of the “thickness” of words complicating (or preventing) thought within the language framework. In a sense it was a failure of language comprehension, but he wasn’t using the framework of language to understand the words at all; rather he was constructing a different but equally valid understanding of the words through a sensory framework. 

Dan Miller, an autistic artist, operates in a very similar way. Though he does not write about his work, and reviewers do not speculate about the particulars of his ways of thinking, it’s clear that he is working in a similar visual framework that is synesthetically tied to sound. His paintings are built-up layers of repeated written words in various colors and textures, forming visual structures that almost resemble abstract landscapes. Similar to Mukhopadhyay, his work deals with words in the context of their physical environment. The words’ forms are built up with colors that, in a synesthetic framework, reflect the vocal properties of tone and register. He also includes forms of basic objects he notices, like electric plugs and lightbulbs, which influences the context of the words and their forms. Overall, the paintings visually represent what could be described as “soundscapes,” exploring the physical forms of words and how those forms interact with the context they are in. Miller’s work, unlike the examples of autistic writers, does not technically deal with language as a framework at all. It deals with the sensory material that language uses as symbols for imposed social meanings, but it does so in a very sophisticated manner through a purely visual sensory framework. He constructs meaning visually in a way that would be equivalent to an essay within a language framework, but because it does not have to translate into a fundamentally different framework, it retains the character of its internal logic.

Overall, these examples of autistic perception interacting with language both confirm Nietzsche’s account and reveal its limits. His observations on language as a social phenomenon based in an idea of common perception served as an “explanation” for why autistic perception results in such a disconnect in understanding within language. It’s revealing that autism is characterized as a communication disorder rather than merely a sensory one; the idea of shared perception is so fundamental to language’s functioning that when it is proven wrong, communication becomes incredibly difficult. Language and sensory experience are inextricably linked, as Nietzsche says, not by the nature of language, but through social construction. However, while his account is correct within the social context of language, autistic modes of thinking show that his ideas on language and perception are not all-encompassing. He is correct that the framework of language that he associates with conscious thought is foreign to sensory perception and is unnaturally “imposed” on it after the fact. But as the examples of autistic sensory frameworks show, thought occurs in more ways and more contexts than linguistic conceptualization. Nietzsche’s explanation does not account for the existence of associative, index-based thought, which is inextricably connected to sensory perception. Not all thought is foreign from experience, and the near-universal autistic experience of sensory thought provides a substantial counterexample to Nietzsche’s equivalating thought with language. This would suggest that while his arguments on language are incredibly relevant, his resulting cynicism about the separation of thought from perception may be premature. Interpretation of his work will benefit from a broadened scope of view that can encompass autistic ways of being and the stunning variety of human experience.

notes

1 Nietzsche 143
2 Ibid. 145
3 Ibid. 144
4 Ibid 146
5 Ibid. 145
6 Ibid. 146
7 Ibid.
8 Ibid. 149
9 Ibid. 150
10 Ibid. 147
11 Hahamy
12 Ibid.
13 Savarese 44
14 Nietzsche 149
15 Nietzsche 144
16 Nietzsche 147
17 Bogdashina 84
18 Bogdashina 85
19 Ibid.
20 Autism and Sensing 99-100
21 Ibid.
22 Grandin 32
23 Bogdashina 86
24 Grandin 19
25 Grandin 37
26 Bencivenga 101
27 Bencivenga 102
28 Bencivenga 103
29 Bencivenga 102
30 Ibid.
31 Note that Bencivenga’s use of this example is further violent imposition onto physical/historical reality: by using
an enslaved person’s pain as a demonstration of a philosophical argument, he is expecting the audience to “see
through” the act of violence from a presumed shared focus on Benciventa’s argument, rather than the enslaved
person’s reality.
32 Ibid.
33 Nobody Nowhere 4
34 Savarese 45

works cited

Bencivenga, Ermanno. My Kantian Ways. University of California Press, 1995. 

Bogdashina, Olga. Autism and the Edges of the Known World. Jessica Kingsley Publishers, 2010.

Grandin, Temple. Thinking in Pictures. Vintage Books, 1995.

Hahamy, A., Behrmann, M. & Malach, R. “The idiosyncratic brain: distortion of spontaneous connectivity patterns in autism spectrum disorder.” Nat Neurosci, vol.8, 2015, pp. 302–309. https://doi.org/10.1038/nn.3919.

Nietzsche, Friedrich. The Birth of Tragedy and Other Writings. New York, Cambridge University Press, 1999.

Savarese, Ralph. See It Feelingly. Duke University Press, 2018. 

Williams, Donna. Nobody Nowhere. HarperCollins Publishers, 1992.

Williams, Donna. Autism and Sensing: The Unlost Instinct. London, Jessica Kingsley Publishers, 1998.