The Holy Priest with the Vulgar Tongue - What the F: What Swearing Reveals About Our Language, Our Brains, and Ourselves - Benjamin K. Bergen

What the F: What Swearing Reveals About Our Language, Our Brains, and Ourselves - Benjamin K. Bergen (2016)

Chapter 4. The Holy Priest with the Vulgar Tongue

Jacques Lordat is the most important neuroscientist you’ve probably never heard of. Born in 1773, he trained as a physician and then practiced medicine and served as a professor in Montpellier, France. But at the age of fifty-two, he suffered a devastating stroke. A stroke, as you might know, occurs when a blocked or leaky blood vessel reduces blood flow to part of the brain. Deprived of the oxygen that the blood carries, neurons in the affected region start to die off, and this can cause long-term impairments to the functions that rely on those cells. Lordat’s stroke apparently stemmed from a tonsil abscess that led to a blockage in the carotid artery,1 which brings blood to the front of the brain. And it left him unable to speak. But slowly and with effort, he began to heal himself (as physicians will). And throughout his recovery, even in the early days when his capacity for speech was decimated, he could still think lucidly. So when he eventually regained the ability to speak and write, he recorded for the scientific record not only the objective facts of his case but also the subjective experience of what it feels like to lose language due to brain insult. And he eventually used his expertise to lay out the first modern theory of how our brains produce language, a theory that’s still broadly recognizable in the contemporary scientific consensus nearly two centuries later.2 So, in short, he’s kind of a big deal.

After his stroke, Lordat focused his research and practice on people who, like him, had suffered language-compromising brain damage. One of his most surprising findings came from a case study, the account of which he published in 1843. In it, he wrote about a parish priest, a man of God, who, like Lordat himself, had suffered a serious stroke.3 The priest also showed obvious signs of language impairment. Just like Lordat, he retained little ability to speak—only more so. The priest’s vocabulary was reduced to just two words. The first was je, the French word for “I.” Perhaps unsurprising, if you’re only going to retain one word. But the other word was, at the time, unimaginable for a parish priest. Lordat wrote that the priest made ample use of “the most forceful oath of the tongue, which begins with an ‘f’ and which our Dictionaries have never dared to print.”4 That was the French word foutre, which by coincidence happens to start with the same letter as its English translation: fuck.

In his own experience and that of the priest, Lordat had discovered aphasia—language impairment caused by damage to specific parts of the brain. Unfortunately, aphasia is common. Currently, about 1 million Americans suffer from some type of aphasia due to brain damage at the hands of stroke, as well as traumatic brain injuries, infections, and dementia.5 And with two centuries of accumulated observations, we now know that many of these people exhibit the very same syndrome that Lordat observed in his priest. They find it difficult or impossible to intentionally articulate and string together words. But some spontaneous language is still preserved—interjections like yeah and huh, filler words like um and well, and some of the most vulgar verbal ejaculations in the language.

How can this be? Why does certain brain damage render its victims unable to articulate even the simplest sentences but leave knee-jerk profanity intact? What does this imply about the brain and how it produces language—whether in sickness or in health?

Aphasia is a cornerstone of neuroscience. It’s the most revealing tool we have to study the brain’s mechanics for generating and understanding language—how something breaks down often provides the clearest window into how it works. And while Lordat’s observations and his theory have been influential in the field that he founded, the theories that have built up over the centuries, based in large part on aphasia research, have largely omitted the profanity. With few exceptions, it’s become an anecdotal side note.6 And as I’ll argue, that has led to exactly the wrong conclusion about how the brain works.

# $ % !

The brain is a roughly three-pound sack of tissue, populated by billions of neurons soaking in a finely tuned chemical bath. The ancient Greeks thought that the brain’s primary function was to cool the body—that it served as a sort of organic radiator.7 In truth, the brain does consume about 20 percent of your body’s energy (despite making up only 2 percent of your body weight).8 But we now know of course that it does more than just produce heat. It also generates light, at least metaphorically speaking, in the form of thought. More relevant for our present purposes, it also allows you to learn, produce, and understand language. When things go right with language, it’s because the brain is working. When language breaks down, it’s usually due to brain malfunctions. For example, when a well-mannered parish priest spontaneously starts cursing a blue streak, you can bet it has something to do with how language is fleshed out in his brain.

The evidence we can glean from aphasia—from Lordat onward—shows that language is implemented in the brain in a highly structured way. Different parts of your brain are involved in different sorts of operations. Known as localization of function, this is one of the most important principles of brain organization.

Here’s how it’s supposed to work for language. Suppose you hear someone utter a word.9 Once the sound hits your ears, the inner ear converts it into electrical signals, pulling apart the sound’s different frequencies like the equalizer on a stereo. This electrical signal bounces through a chain of specialized areas in your brain until it reaches the auditory cortex in the temporal lobe, which processes sounds in general. Different parts of the temporal lobe extract information about the speech sounds that make up the word and then send modified signals to a region called Wernicke’s area, which is believed to associate the sequences of sounds that you’ve heard with their meanings. It’s thought to be like a mental lexicon. (If the word you’ve just heard is embedded in a sentence, then there’s more work to do, involving other brain areas, but let’s leave that out for now.) That’s how comprehension is supposed to work. Conversely, to produce language, you might start with Wernicke’s area, which could allow you to select words that will adequately express whatever meaning you want to convey. And then signals get transmitted up to Broca’s area in the front of the brain, where they’re translated into sounds to be articulated.

The evidence that these different areas serve these particular functions—that functions are “localized” in this way—comes originally and most compellingly from aphasia. Damage to the different parts of the brain produces distinct aphasia. When Wernicke’s area is damaged, people have trouble understanding language. Moreover, when they talk, they pronounce the words and sentences fine; they just don’t make sense. For instance, someone with Wernicke’s aphasia might say something like “I did the thing in the thing.” It’s a perfectly grammatical sentence, but what does it mean? Maybe the speaker wants to tell us she washed the dishes. Or maybe that she seduced an orderly. Who knows?

A brain (schematically, with skull removed). This is the outside of the left hemisphere of the brain—the front of the brain is on the left-hand side of the page, and the top is up. Modified from an image licensed under the Creative Commons Attribution—ShareAlike 3.0 Unported License. Source: Hugh Guiney

A brain (schematically, with skull removed). This is the outside of the left hemisphere of the brain—the front of the brain is on the left-hand side of the page, and the top is up. Modified from an image licensed under the Creative Commons Attribution—ShareAlike 3.0 Unported License. Source: Hugh Guiney.

A Broca’s aphasic, on the other hand, uses the right words, and you can usually figure out what he means; it’s just that the structure of his words or sentences (or both) is off. He will also have trouble pronouncing speech sounds and the words they make up. So a typical utterance might be “I-I-I make … um … damn!” With effort, you can sometimes figure out what he’s getting at, but the articulation of words is labored. Because Broca’s area is often damaged in people who show the symptomatology of Broca’s aphasia, it follows that this region subserves the capacities lost in Broca’s aphasia. Both Lordat and his priest suffered from devastating damage to this part of their brains—to the point where almost all production of speech was obliterated.

This is a remarkably clean story. If it’s right, specific brain regions perform different functions for language—Wernicke’s computes meaning, and Broca’s deals with sound. This is the prevailing view of how the brain processes language that you’ll see in most introductory cognitive psychology or cognitive neuroscience textbooks (like the one in the footnote at the end of this sentence).10 And the idea that different parts of the brain do different things (localization of function) is appealing. It’s as if the brain houses an efficient assembly line for language.

The problem is that the case of the cursing priest doesn’t square with this story. When people suffer from Broca’s aphasia, the condition doesn’t affect all words equally. It plays favorites. Broca’s aphasics often struggle to articulate or remember how to produce run-of-the-mill nouns and verbs. But fixed expressions like How do you do or I can’t, and especially expletives, tend to be preserved. So it’s not uncommon to find the language of Broca’s aphasics, like Lordat’s priest, liberally sprinkled with Jesus Christs, shits, and motherfuckers.11 Even when patients suffer from nearly complete loss of language—so-called global aphasia—they often retain interjections and profanity. For instance, a recently documented patient could only produce six words: well, yeah, yes, no, goddamnit, and shit.12

This same pattern, which shows up in patient after patient, requires a fundamental rethinking of the brain’s functional organization for language. It can’t be that all language is produced on the same assembly line. Just as neuroscientists have concluded that meaning and articulation are localized separately in Broca’s and Wernicke’s areas because each can be selectively impaired by damage to the respective regions, so differential impairment of profanity versus the rest of language implies that these too must have different neural bases. The question then becomes, if the parts of the brain that we think perform language functions (like Broca’s and Wernicke’s) aren’t responsible for a frustrated fuck, goddamnit, and shit, then what is?

We uncover clues when we dig a little deeper into what aphasics can and can’t say. It turns out that the patient mentioned above who could only say six words (and possibly the priest as well, although Lordat’s writings don’t record this) could only say his six words reflexively. That is, he could only produce shit as an unintentional reaction—a fleeting expletive—but he found it impossible to read or repeat the word shit when asked to do so intentionally.

So while it would be noteworthy if we were to discover that profanity is simply generated by different parts of the brain than nonprofanity, there’s actually something far more revealing going on. The issue isn’t about which words are produced so much as how they are used. It seems that while Broca’s aphasia and global aphasia impair intentional speech, they don’t interfere with speech that is reactive, impulsive, and spontaneous—so-called automatic speech.13 This distinction is crucial for understanding how the brain produces language. Automatic speech dissociates from intentional speech—they can be differentially impaired. This means that they must originate in different mechanisms of the brain. A word, when used as an automatic expletive, is generated by one set of brain circuitry; but when crafted intentionally, that same word comes off a different assembly line. Accepting for the time being that Broca’s and Wernicke’s areas bear much of the burden for intentional speech, we’re left to ask, where in the brain do the fleeting expletives of automatic speech come from?

# $ % !

As you likely know, the human brain is split down the middle into two rounded, roughly symmetrical halves—its two hemispheres. These two hemispheres do slightly different things. You may have heard of people who are “right-brained” being more artistic and people who are more “left-brained” being more logical. That’s mostly hooey, but like a lot of hooey, it’s based on a grain of truth.14 Namely, while everyone with two normal hemispheres uses both of them in parallel to do pretty much anything, from crossword puzzles to rhythmic gymnastics, it is true that your brain does some things in a partly lateralized fashion. That is, some functions use one half of the brain more than the other. And some aspects of language are like this.

In most people, the traditional language centers—Wernicke’s area, Broca’s area, and so on—are all clustered in one half of the brain. In almost all right-handed people (95 percent), the left hemisphere is more active during language tasks than the right hemisphere. Being left-handed makes it more likely that the reverse will be true—that the right hemisphere will outwork the left, but this is still only true in 27 percent of left-handers.15

It might stand to reason, then, that automatic language, like the rest of language, would tend to be produced using the left hemisphere more than the right hemisphere, and more so in righties than in lefties. And the dissociation logic would imply that because automatic speech can be preserved even when Broca’s and Wernicke’s areas are damaged, automatic language might be generated by some other part of the left hemisphere, outside of the well-known language centers.

A fascinating little bit of evidence on this comes again from people with aphasia. It involves looking not directly at their brains or what words they’re able to say but rather at how they use their mouths to articulate words. For this to make sense, you have to know something slightly counterintuitive about how the brain controls physical actions. The left hemisphere of the brain is primarily responsible for sending signals to muscles on the right side of the body—the “contralateral” side—and the right hemisphere of the brain controls actions on the left side. This contralateral control includes moving your hands and feet, and it also includes moving your mouth; the left hemisphere takes the lead in controlling the right side of your mouth, and the right hemisphere is in charge of the left side. Because most people’s brains are predominantly left-lateralized for language, they (and that probably includes you) actually speak using the right side of the mouth slightly more than the left side.16 So when you open your mouth to say a word, the right side tends to open slightly wider than the left side. And it’s not just that you’re leaving the right side of your mouth slightly more agape overall—the total distance moved on the right side is farther than on the left side.

But that’s not true of everything you do with your mouth. On the next page you see what happens when you measure mouth lateralization as people (in this case, people with Broca’s aphasia and damage to the left hemisphere of the brain) perform a variety of language tasks. Some are purely linguistic, like generating lists of words or repeating words; others are less so, like singing and smiling. The researchers measured the relative aperture of the left and right sides of the patients’ mouths as they moved them in all these different ways. And what you see is striking. These people opened the right side of their mouths more for most language tasks—generating lists of words, repeating words and sentences, saying whatever came to mind, and so on. This indicates more left-hemisphere control. But automatic speech (labeled here as “serial speech”)—things like counting from one to ten or reciting the alphabet—induced larger apertures on the left side of the mouth, suggesting greater right-hemisphere control. Like swearing, counting and recitation are often preserved in Broca’s aphasia, which is why they’re considered “automatic speech.”

The relative opening of the left versus the right side of the mouth indicates which hemisphere of the brain is doing more work in controlling speech, and these results show that while most speech is left-lateralized in the brain (because the right side of the mouth is open more), automatic speech is right-lateralized. Now, the researchers didn’t elicit spontaneous expletives. But if these pattern with other automatic speech, they might actually have a right-hemisphere origin.

Relative frequency of larger openings on left and right side of the mouth in twenty Broca’s aphasics with left-hemisphere impairment. The right side is opened wider during all language tasks except singing and automatic language. Produced on the basis of data in R. Graves and T. Landis (1985)

Relative frequency of larger openings on left and right side of the mouth in twenty Broca’s aphasics with left-hemisphere impairment. The right side is opened wider during all language tasks except singing and automatic language. Produced on the basis of data in R. Graves and T. Landis (1985).

How can we pursue this idea? The best type of evidence for the lateralization of spontaneous profanity to one hemisphere or the other would come from people who can only use one brain hemisphere. If those people can use one type of language—automatic or intentional—but not the other, this would suggest that the nonworking hemisphere is necessary for the impaired type of language. And hard to believe though it might be, there are in fact people who can use only one hemisphere of their brains—because one hemisphere is all they have. These are patients who have had one hemisphere partially or completely removed for medical reasons.17

One such individual was a patient pseudonymized as “E.C.,” a forty-seven-year-old, right-handed man who developed severe symptoms from a substantial tumor in the left hemisphere of his brain. As you might expect, since each hemisphere of the brain is the primary driver of motor actions on the opposing side of the body, these included not only aphasia but also motor deficits specifically affecting his right hand. He was admitted to the hospital in March 1965, and the tumor was removed from the left hemisphere of his brain. But the tumor turned out to be malignant, and as E.C.’s symptoms did not improve after surgery, the decision was made in December of the same year to remove the entire left hemisphere of his brain. Given how radical this procedure is—removal of an entire half of the brain—his prospects must have been bleak.

So here we have someone who—if language is left-lateralized—should have no ability to speak. And indeed, after surgery, most of his language capacities had been obliterated. But not all. A report of his condition reads as follows: “E.C.’s attempts to reply to questions immediately after operation were totally unsuccessful. He would open his mouth and utter isolated words, and after apparently struggling to organize words for meaningful speech, recognized his inability and would utter expletives or short emotional phrases (e.g., ‘Goddamit!’). Expletives and curses were well articulated and clearly understandable.”18 Let me just highlight the important bits. He was totally unable to produce meaningful speech. Then he got frustrated and swore: “Expletives and curses were well articulated and clearly understandable.” Just as a reminder, this person was missing the entire left side of his brain. This is an astounding discovery. You don’t need your left hemisphere to talk, as long as you’re swearing in frustration. It seems that part of the reason that Broca’s and even global aphasics still swear is that circuitry located in the right hemisphere can generate automatic speech.

What is that circuitry like? Are there right-hemisphere homologues of Broca’s and Wernicke’s areas, specialized for automatic speech? Is it some different type of machinery entirely?

As we narrow the scope of the question from which hemisphere is involved to which regions within that hemisphere, the type of evidence we need also becomes more specific. The critical evidence would have to come in the form of people who have suffered damage to different parts of the right hemisphere of their brain. You look to see whether any of them has trouble only with automatic speech but not with the rest of language—basically, the opposite of Lordat’s priest. And you use their language dysfunction to home in on the regions in the brain that house the circuits that generate automatic speech. So the challenge becomes finding people who, due to localized right-hemisphere brain damage, lose the ability to spontaneously produce epithets.

In the literature on aphasias, there are no such people. At least not “people” in the plural. There’s just a person. One of them. I present for your consideration that one case. A 1993 report in the journal Neurology describes a patient who—like many—had damage to the right hemisphere of his brain.19 But the behavioral consequences of this damage manifested in a peculiar way. This patient, who was bilingual in French and Hebrew, spoke like a typical speaker of both of his languages. That is, until it came to automatic speech. Reportedly, after suffering brain damage, he was unable to sing familiar songs or recite nursery rhymes, and he couldn’t spontaneously swear. And this was most surprising because before suffering brain damage, he was reportedly an enthusiastic purveyor of profanity. Here he is: the mirror image of the parish priest! Instead of preserving automatic language, such as epithets, his brain damage specifically impaired that function alone.

And now for the payoff: Where was his right hemisphere damaged? What brain circuitry is necessary for automatic speech? The basal ganglia, a system of subcortical brain structures—“subcortical” because they lie embedded beneath the cerebral cortex. They play a role in selecting appropriate motor actions by inhibiting ones you don’t want to perform and are closely tied to emotion centers of the brain.20 This patient lost the ability to blurt out emotionally charged idioms when brain damage compromised the functioning of his right-hemisphere basal ganglia.

The basal ganglia, nestled underneath the cerebral cortex, are involved in selecting and inhibiting motor actions. Source: Modified from an image by John Henkel, of the Food and Drug Administration

The basal ganglia, nestled underneath the cerebral cortex, are involved in selecting and inhibiting motor actions. Source: Modified from an image by John Henkel, of the Food and Drug Administration.

From a single case study, we wouldn’t want to jump to the conclusion that the basal ganglia are always necessary for automatic speech. We also don’t know what aspect of automatic speech they’re responsible for or whether they’re necessary for all types of automatic speech. But this one case suggests that automatic language—including spontaneous swearing—might be generated by brain circuits responsible for automatic processes other than language that are tightly linked to emotion centers. We’ll explore each of these themes in the remainder of this chapter. But as an intermediate conclusion, we’ll have to content ourselves with this: automatic swearing is localized differently from other types of language, and as a result it behaves differently when the system is stressed or damaged. Although specific brain areas like Wernicke’s and Broca’s are vital to using a lot of language, many others are also involved. Language is manifested heterogeneously in the brain.

# $ % !

The evidence from automatic aphasia implicates the basal ganglia in automatic speech. On its own, this evidence would be suggestive but not conclusive. As it turns out though, the basal ganglia can come to have altered functioning in other ways. Tourette’s syndrome is a hereditary neurological disorder affecting the basal ganglia,21 which, in line with these structures’ known functions, affects people’s ability to control their own actions. Tourette’s is characterized by the production of spontaneous and uncontrollable tics. These tics can take the form of movements of the body or face or undecipherable noises that sound like grunts or throat clearing. Or the tics can be actual words.

In some cases, the words that people with Tourette’s uncontrollably utter are taboo. This symptom, possibly the best-known aspect of the syndrome, is called coprolalia (that’s from the Greek kopros [“feces”] and lalia [“speech”]). Coprolalia is the feature of Tourette’s focused on most by the media—see, for instance, the depiction of a person with Tourette’s by Amy Poehler in the 1999 film Deuce Bigalo: Male Gigolo. The reasons for this fascination are perhaps obvious: losing control over taboo language is more shocking than losing control over clearing your throat. But despite this attention to the taboo side of the syndrome, only somewhere between 10 and 50 percent of people with Tourette’s (depending on who’s counting and whom they count) actually display coprolalia.a Nevertheless, coprolalia has been documented in people with Tourette’s speaking dozens of languages, including English, Japanese, Czech, and even signed languages, where signers uncontrollably produce obscene signs.b Some hearing people with Tourette’s also make obscene gestures uncontrollably (so-called copropraxia).22

I know what you’re thinking. If Japanese doesn’t have profanity, then how can a Japanese speaker have coprolalia? According to the three case studies I know of (summarized in Van Lancker and Cummings [1999]), Japanese coprolalia includes insults (like baka [“foolish”] or dobusu [“ugly”]) and childlike words for genitalia (like chin-chin [“penis”]). They’re words that describe taboo concepts or that you wouldn’t use in polite situations, even though the words themselves aren’t profane.

The difficulty in assessing the true incidence has several causes. Most critically, the diagnostic criteria for Tourette’s syndrome have changed over the years, most notably in 2000, when the DSM-IV loosened them to include any person experiencing persistent vocal and motor tics (removing age-of-onset and frequency criteria, among others). This led to increased diagnosis of Tourette’s syndrome and a resulting decrease in coprolalia proportion estimates.

For the people it affects, living with coprolalia can be challenging. Imagine being unable to restrain yourself from uttering profanity in the supermarket or the office or when picking your child up from kindergarten. Some people with coprolalia describe the compulsion as akin to a sneeze—you feel a growing impulse that can only be alleviated by its release. Because it’s hard to suppress coprolalia—just like sneezes—people with this condition often take the path of avoidance, finding ways to stay out of public as much as possible. They often also develop coping strategies for those occasions when other people are around. One of the most effective is to replace words or parts of words when they burst out. Apparently, it’s harder for many people with coprolalia to suppress or substitute a whole word than to modify just the end of a word that’s coming out anyway. For example, if a person’s impulse is to say cock, closing the mouth intentionally toward the end of the word will produce cop instead. It might seem strange to run around yelling cop, but it’s hardly as socially stigmatized as the alternative.23

If we set coprolalia and the swearing preserved in aphasia side by side, we can see that they differ in revealing ways. For instance, nothing like the substitution strategy has ever been described in aphasia, and aphasics don’t feel the welling up of an urge to curse experienced by most people with coprolalia. But more interestingly, the specific taboo words selected are mostly distinct. Aphasics with preserved automatic speech often produce predominantly expletives expressing frustration or surprise, like shit, fuck, or goddamnit. But the expletives present in coprolalia tend toward words for body parts and bodily effluvia, as well as racial, ethnic, and gender-based slurs. This is true across languages.24

Both types of profane words—those preserved in aphasia and those that burst out in coprolalia—express strong, transient emotional states. This has led some theorists to propose that swearing of both the aphasia and the coprolalia types is produced by different brain machinery than the rest of language.25 As I mentioned earlier, it’s possible that there’s one pathway for producing a lot of language—the one that’s been principally studied in humans and that in most people passes predominantly through the language centers of the left cerebral cortex and is used for the systematic, intentional composition of normal language. The second purported pathway is evolutionarily far older and shared with other mammals who themselves are bereft of anything like human language.

The limbic system, emotion-generating regions deep in the brain, dominate this proposed second circuit. The basal ganglia are directly adjacent to and tightly interconnected with brain structures that process emotions, like the anterior cingulate, the hippocampus, and the amygdala. These ancient brain structures appear to play a role in generating emotional states that create motor impulses, which the basal ganglia then have to regulate and selectively suppress. In the case of coprolalia, the compromised basal ganglia are unable to suppress verbal impulses along this pathway, which results in the characteristic expletives.

Work with other animals—particularly other primates—has revealed closely homologous circuits. For instance, when neurons in the limbic system of macaques or squirrel monkeys are stimulated, the animals spontaneously produce emotional vocalizations.26 This implies that in the typical human brain, perhaps even mine and yours, everyday impulsive, automatic, emotional swearing may be driven by this very same circuitry—circuitry that is a mammalian or primate innovation rather than a uniquely human one. This circuit fills a vital evolutionary function for social beings, allowing an individual to transmit a signal identifying its internal emotional state readily and efficiently to conspecifics. If analogous circuitry is indeed responsible for reflexive human swearing, then it provides privileged access to emotion in the brain, laying bare a speaker’s covert internal experiences unmediated by rational and deliberate planning.

But there’s a caveat. This older, emotion-driven circuit doesn’t behave the same way in humans as it does in other animals. As both Timothy Jay and Steven Pinker have pointed out, the vocalizations we produce when spontaneously swearing are conventionalized—they’re the product of socially driven learning.27 Swearwords are a different beast from shrieks or growls in that they have a specific learned form—you swear specifically in English or Chinese or ASL, whereas a monkey just shrieks in Monkey.

# $ % !

The ramifications of these brain facts are manifold. First, it appears that the classic view of language as left-lateralized and cortical, subserved by a set of distinct and specialized brain circuits (e.g., Broca’s and Wernicke’s areas), tells only part of the story. We also use another pathway for language, built from machinery that’s far older in evolutionary terms and shares little brain circuitry. Language is not monolithic in the brain.

And on a bit of reflection, this makes sense. We use language for multifarious purposes. We should hardly expect the brain systems that support the production of anodyne phrases articulating rational thoughts (like the present sentence) to have the capacity to do double duty, also connecting abrupt, hot, emotional states like frustration and rage to the spontaneous, primal utterances that express them.

And even this separation of the language capacity into two pathways surely understates the variety present in the brain. Although our evidence on the issue is poor at present, it wouldn’t be shocking if we used distinct machinery for the various things we do with language. Do we use a separate pathway for conventionalized greetings (Hi. How are you? Fine.)? Do we use another for onomatopoeia—words like cock-a-doodle-doo that sound like what they mean? And is more detailed variation afoot even within the two ostensible pathways we’ve been discussing? Is our two-way distinction too coarse? Perhaps we recruit distinct pathways when we spontaneously express frustration versus anger, fear versus arousal. To date, we just don’t know, in large part because the low-level neuroscience depends on having animal models to work with, and other animals share some but not all of our brain circuitry: Broca’s and Wernicke’s areas appear to be largely human specific, for instance. And other animals display some but not all of the communicative functions we deploy language for. In the future, better imaging techniques applied to the functioning human brain will surely reveal the extent of the diversity of neural instantiations of language behavior.

Finally, it’s also worth reiterating here that it’s impossible to equate different brain pathways with different words. We can produce the same words in different ways. When compromised brain function leads to language deficits, it’s usually not specific words but specific ways of using them that are lost. Broca’s aphasia usually impairs deliberate, intentional articulation of words and preserves some automatic speech. That is, even an aphasic priest who spontaneously and fluently produces “the most forceful oath of the tongue” when frustrated will be unable to intentionally articulate the very same word.28