Features of Language
There is no human society that lacks language, and, therefore, its presence is a cultural universal. Is that where the universality ends?
The 7,000 or so languages in the world are wonderfully diverse, with different sounds, words, and rules for making sentences. The !Kung language spoken by the San of the Kalahari, for instance, has one of the largest sound inventories in the world including distinctive click sounds. Different languages use different rules for ordering subjects (S), objects (O), and verbs (V). In English, we say “Sue sells salsa”, an SVO sentence. In Tuareg-Berber, a VSO language, the sentence would be something like “Sells Sue salsa.” There are 6 different possible arrangements for sentences, all of which are found in human languages. Though we often think of language as being spoken, it can have different delivery systems including speech, sign language, and writing. While all cultures have some form of spoken language and accompanying oral traditions, not all have a written form of language. Writing itself had its beginnings around 6,000 years ago, the earliest known writing system, cuneiform, arising in Mesopotamia.
It is easy to take language for granted; most of the time we don’t even think about it. We just do it. If we had to explain it to alien visitors or even to children, we’d probably have trouble even with the basics. We might start with a biological or mental explanation like we saw previously. But we know from our own experience with language every day that it is more than a purely biological and mental process. It’s cultural! In 1960, linguist Charles Hockett defined the features of language. He argued that if a communication system lacks even one feature, it is not language. Because language is so essential to human culture, peeling back the layers of language is useful for understanding the nature of being human. Let’s look at seven of those features:
- Cultural Transmission: humans learn the language of their environment. It is not too surprising that children raised hearing Swahili all around them will grow up to speak Swahili.
- Arbitrariness: our words don’t necessarily match the things they represent. Arbitrariness is what allows different languages to have words that sound nothing like each other. We have all just decided (‘agreed’) on their possible meanings. And sometimes (often, in fact) this means we can get creative with our words. Consider, for example, the new meaning “giving” has taken on even just in the past couple of years.
- Speaking of which, Productivity: new words or sentences can be generated. We even have a word for newly coined words: neologism. Productivity also means you can understand a sentence even if you’ve never experienced that particular collection of words before.
- In part because we can refer to things in a variety of ways, Displacement means we can communicate ideas that are not in our immediate environment or that exist only in our imaginations.
- Discreteness: every human language is made up of a small number of meaningless discrete sounds. That is, the sounds can be isolated from each other, for purposes of study by linguists, or to be represented in a writing system. (A suddenly added new sound would take time to 'catch on.')
- Grammar: the structural rules of human language (even if they are inconsistently used). Rules can be broken, but certain patterns such as how to turn nouns into verbs, for example, help keep the arbitrariness and productivity in check. Grammar and other sorts of linguistic rules are something to ‘fall back on’ as we construct new language and helps ensure we continue to understand one another.
- And, finally, language is Context-specific. Language doesn’t work in isolation.
There are, of course, many other ways that animals communicate with one another. But, so far as we can tell, humans are the only creature with ‘language’ holding all of these characteristics. In this chapter, we’ll focus on the social and cultural implications of language and communication--and language change. Some anthropologists (such as Edward Sapir and Benjamin Lee Whorf) have even gone so far as to argue that all culture is linguistic (e.g., when we say that an object has ‘meaning,’ we mean it contains a form of oral history). This might be extreme, but, at the very least, anthropologists do care how people organize their language, and you can learn a lot about a society by studying how it communicates.
Hidden in What We Say
Language is a fundamental aspect of human culture and a key element in constructing social identity. In anthropology, the study of language overlaps with a wide range of topics, from the structural aspects of different languages to the cultural contexts in which they are used—pretty much every one of Hockett’s features above. Here are just a few examples I like that help show why language is important in comparative anthropology: Shibboleths (which we've seen before), the distinction between orality and literacy, and the potential for universals (or not).
Shibboleths
Shibboleths get their name from a biblical story which serves as a pretty obvious example of what they are. A Shibboleth is a word or phrase that is used to distinguish members of one group from those of another. A modern example might be whether you say ‘bubbler’ / ‘drinking fountain,’ ‘traffic lights’ / ‘stop and go lights,’ ‘casserole’ / ‘hot dish,’ how or if you pronounce the first T in ‘interesting,’ or how you pronounce ‘pecan’ or ‘caramel.’ The term originates from the biblical story in which the Gileadites used the word “shibboleth” to identify and kill members of the Ephraimite tribe, who could not pronounce the word ‘correctly’ according to their culture. The two tribes were enculturated to pronounce S and SH sounds differently. In a broader sense, Shibboleths are linguistic markers that signal social, ethnic, or regional identity. For linguists today, these linguistic markers are not just about pronunciation; they can also include vocabulary, grammar, and even non-verbal communication cues.
Here are some examples:
- Dialectal differences: Many Aboriginal Australian languages have regional variations, or dialects, that can be used to identify where someone is from. There are more than 250 indigenous Australian languages including nearly 800 dialects.
- Click languages: The Khoisan languages of southern Africa are known for their distinctive click sounds. The distribution and ‘type’ of clicks used can distinguish Khoisan speakers from speakers of other African languages.
- Tonal languages: East Asia and Africa have more ‘tonal’ languages than the rest of the world. The ability to correctly use tones can be a marker of language proficiency and cultural identity—especially considering the difficulty of learning a tonal language for a Western (largely non-tonal) speaker. The closest example of tonal language in English, for example, is the difference in inflection between a statement and the same statement as a question. Note the difference in tone between: “You’re wearing pajamas.” and “You’re wearing pajamas?”
- Honorifics: Many Southeast Asian languages have elaborate systems of honorifics, which are used to show respect to others. The correct use of honorifics can be a sign of social status and cultural knowledge.
- Dialects: Most languages have dialects. Native American languages, for example, are known for their regional dialects, which can be used to identify where someone is from. Speakers of Navajo from the Four Corners region may have a different dialect than speakers from the Navajo Nation’s eastern border. With the steady decline in number of speakers, these dialectical differences are under even more threat of extinction.
- This idea of a cultural form being associated with ‘us’ and ‘them’ should not be new to you. In fact, we saw this in research on the specific way that speech is organized in terms of social groups—especially between genders.
Part of the reason ideas like Shibboleths are so powerful is because, from its very basis, language is about the human drive to classify things (including themselves), both making them (or saying they are) similar and to say they are different. Language is never just neutral because it is always about how people relate to each other. Even the most innocuous language tells us about who/what gets included or excluded. We can see this in as simple as how a culture divides up the spectrum of visible light into named colors or as complex as life stages, a spiritual hierarchy, or caste system. Consider, too, the ‘meta-language’ of what is ‘said’ by the rules of who gets to speak, to whom, and when.
Orality and Literacy
Next, we turn to the distinction between orality and literacy. Orality refers to cultures that primarily rely on spoken language for communication, while literacy refers to cultures that use written language. This distinction is one we probably don’t consider too often in the modern West where everything is written about. But this distinction is crucial in understanding how different societies transmit knowledge, maintain social structures, and preserve cultural heritage.
In oral cultures, storytelling, proverbs, and oral traditions play a central role. For instance, among the Yoruba people of Nigeria, oral literature, including folktales and praise poetry, is a vital part of cultural expression and social cohesion. In contrast, literate cultures often rely on written texts to record history, laws, and religious practices. The ancient Egyptians, for example, used hieroglyphics to document everything from administrative records to religious texts. Babylonians also had a form of writing. But not all writing is created equal—something modern archaeologists have to contend with when they try to ascribe modern writing practices to ancient systems.
The boundary between orality and literacy is not always clear-cut. Many cultures exhibit a blend of both. The Maori of New Zealand, for instance, have a rich oral tradition that includes the recitation of genealogies and myths, but they also adopted writing systems to preserve their language and culture in the face of colonial pressures. Similarly, the Mayan civilization of Mesoamerica developed a complex system of syllabic hieroglyphic writing while maintaining (we think) strong oral traditions that went along with them.
What is important about these ways (oral, literal, or something else) of doing language is that they can change over time (in part because of those seven features outlined above). More importantly, what we think of as “natural,” then, is actually a product of a particular historical and cultural period.
I also probably don’t need to list even one example of how language is also tied to emotions. How do you think this link between words and sentiments is different (or the same) between the spoken and the written word? Studies suggest that the emotional tone of language make it easier for people to remember situations or events that they found disturbing, joyful, angry, etc. One reason, perhaps, that rituals are effective is that when people cry, laugh, etc. they are “buying in” to the commitment involved. And this idea of social commitment and memory brings us to a more detailed account of Orality and Literacy.
Orality
Orality is simply speaking, particularly when literacy is largely absent. Walter Ong is best known for his theorizing on the implications of orality along with Milman Parry and Albert Lord who focused on oral performances such as story-telling and epic poems.
Ong argued that oral cultures live in a more intimate “hearing” environment. This impacts our distance from one another—we need to remain close to hear. As a result, oral cultures have an upward limit in terms of size—other historians and archaeologists have made a similar point. Beyond a certain population size, you can no longer (reliably) hear everything that happens. In literacy, by contrast, communication shifts from the ear to the eye—and the network with which you communicate grows tremendously. As a result, literacy is a kind of distancing mechanism that separates people from one another in time and space.
Think about how continuity happens over generations in oral cultures since spoken language goes out of existence as it is spoken. Parry and Lord have an answer. They studied the performances of Serbian folk singers, who might repeat tens or even hundreds of thousands of lines of poetry over a period of several days. Many other oral cultures have lengthy performances as well: the ancient Hebrews undoubtedly performed the Pentateuch; Muslims repeat the Qur’an orally; ancient Norse, English, and Celtic peoples had their epic poems, as did the ancient Greeks; Navaho doctors sing songs such as Blessing Way for four days without interruption.
How? It’s not just about memory. In most cases, these performances follow certain formulas and rules that helped guide them. Since they weren’t written down, performers also had the freedom to adapt their tales to suit the audience. They could recycle phrases and make adjustments on the fly. Who would know—so long as what they said matched up with cultural expectations of the tale! So oral culture is stable because it is performed. How are literate cultures different?
Literacy
Jack Goody has written extensively about literacy. He saw literacy as a social technology that had important cognitive and cultural consequences. Goody points out that prior to literacy, people could not “know” whether one account of an event was the same as another, because there was no way to compare them. As a result, “history” begins with literacy. Science and mathematics, according to Goody, also require writing.
Goody (like Max Weber) saw (Western) government as based on literacy: legal proceedings, taxes, and bureaucratic record-keeping all depend on the availability of written materials. The development of government in its turn made possible the extension of political systems over thousands of miles and hundreds of years. As a result, literacy is often very ‘conservative’—in the sense that it prefers to maintain the status quo with less and less regard for new contexts—since what you are reading may have been written a long time ago and far away.
That said, we always have to take into account the social context of reading and writing in order to understand its effect in specific cultures. For example, until fairly recently, it was considered entirely appropriate to read aloud. The idea that you would read “in your head” and not speak or move your lips is part of the modern emphasis on the importance of individualism—and therefore private personal experience.
Secondary Orality
But, wait. What about YouTube, TikTok, and other forms of 'modern media?' It turns out, societies that go from orality to literacy come to feel the effects of losing the closeness created by hearing and a few select elders and the community at large holding the knowledge (instead of disparate texts).
Ong wrote about what he called “secondary orality” as a form of communication in cultures with mass media. He noted that TV and movies (the mass media of his time) were a lot like the performances of oral cultures, since they tried to establish an intimate space of hearing intent on an emotional response. At the same time, mass media are like literacy because they are based on distance in time and space. “Secondary orality” is a combination of orality and literacy--or a form of literacy trying to be as much like orality as it can be.
Consider the following...
Do you think mass media has erased the physical barriers characteristic of primary orality? Before modern media, various gatekeepers (elders for orality or specialists—such as bureaucrats or librarians—for literacy) controlled access to information. Who are the gatekeepers of mass media like TikTok or YouTube?
Hidden in What We Say
Orality and literacy is one way we can see how drastically different two languages or 'modes' of language can be from one another. Looking at language change can often be about what ‘language’ (as a broad cultural form) says in a culture beyond the words being uttered (or written). Consider what the culture bothers to name, the ways they speak—and if certain classes of people speak differently to one another.
The term anthropologists Deborah Tannen uses for these features language carries with it in addition to the actual content of the language is “metamessage.” The metamessage is whether the speaker is an authority, ‘one of us,’ and so on. A great way to think about metamessage is in ‘code-switching.’ Do you speak to your parents the same way you speak with your friends? To your partner the same way you speak with your boss? It probably wouldn’t take you long to think of a situation in which you or someone else was judged because of the way they spoke.
What many examples the world over all demonstrate is that, while language is a universal human trait, the ways in which it is expressed vary greatly across cultures. What does not change across these variations seems to be that language, like culture itself, is always embedded with a set of beliefs about how the world ‘is’ or ‘ought to be.’
We have previously heard about Keith Basso's “Wisdom Sits in Places.” In it, he looks at the connection between language, landscape, and identity among the Western Apache. Basso illustrates how specific places in the Apache landscape are imbued with cultural significance and tied to stories, moral lessons, and social values that are passed down through generations. They are so much so that the names of places have become a sort of short-hand for certain situations.
The Apache example shows that language is more than just a tool for communication as a central to the way people understand their world and themselves. The Apache use place names not only to navigate their physical landscape but also to connect individuals to their community’s history and moral universe. By invoking place-names, they evoke shared stories and values, reinforcing their cultural identity. These stories, often tied to the moral failings or virtues of ancestors, become a way of teaching lessons about proper behavior, personal responsibility, and community cohesion. Language here acts as a bridge between the physical landscape and the internal sense of self and belonging within a community.
Linguistic Relativity: The Sapir-Whorf Hypothesis
We have already touched on the Sapir-Whorf hypothesis in the previous chapter. In the 1920s, Benjamin Whorf was a graduate student studying with linguist Edward Sapir at Yale University in New Haven, Connecticut. Sapir, considered the father of American linguistic anthropology, was responsible for documenting and recording the languages and cultures of many Native American tribes, which were disappearing at an alarming rate. This was due primarily to the deliberate efforts of the United States government to force Native Americans to assimilate into the Euro-American culture. Sapir and his predecessors were well aware of the close relationship between culture and language because each culture is reflected in and influences its language. Anthropologists need to learn the language of the culture they are studying in order to understand the world view of its speakers. Whorf believed that the reverse is also true, that a language affects culture as well, by actually influencing how its speakers think. His hypothesis proposes that the words and the structures of a language influence how its speakers think about the world, how they behave, and ultimately the culture itself. Simply stated, Whorf believed that human beings see the world the way they do because the specific languages they speak influence them to do so. (Recall examples of naming colors or classifying animals.) He developed this idea through both his work with Sapir and his work as a chemical engineer for the Hartford Insurance Company investigating the causes of fires.
One of his cases while working for the insurance company was a fire at a business where there were a number of gasoline drums. Those that contained gasoline were surrounded by signs warning employees to be cautious around them and to avoid smoking near them. The workers were always careful around those drums. On the other hand, empty gasoline drums were stored in another area, but employees were more careless there. Someone tossed a cigarette or lighted match into one of the “empty” drums, it went up in flames, and started a fire that burned the business to the ground. Whorf theorized that the meaning of the word empty implied to the worker that “nothing” was there to be cautious about so the worker behaved accordingly. Unfortunately, an “empty” gasoline drum may still contain fumes, which are more flammable than the liquid itself.
Whorf’s studies at Yale involved working with Native American languages, including Hopi. The Hopi language is quite different from English, in many ways. For example, let’s look at how the Hopi language deals with time. Western languages (and cultures) view time as a flowing river in which we are being carried continuously away from a past, through the present, and into a future. Our verb systems reflect that concept with specific tenses for past, present, and future. We think of this concept of time as universal, that all humans see it the same way. A Hopi speaker has very different ideas and the structure of their language both reflects and shapes the way they think about time. The Hopi language has no present, past, or future tense. Instead, it divides the world into what Whorf called the manifested and unmanifest domains. The manifested domain deals with the physical universe, including the present, the immediate past and future; the verb system uses the same basic structure for all of them. The unmanifest domain involves the remote past and the future, as well as the world of desires, thought, and life forces. The set of verb forms dealing with this domain are consistent for all of these areas, and are different from the manifested ones. Also, there are no words for hours, minutes, or days of the week.
Native Hopi speakers often had great difficulty adapting to life in the English speaking world when it came to being “on time” for work or other events. It is simply not how they had been conditioned to behave with respect to time in their Hopi world, which followed the phases of the moon and the movements of the sun. In a book about the Abenaki who lived in Vermont in the mid-1800s, Trudy Ann Parker described their concept of time, which very much resembled that of the Hopi and many of the other Native American tribes.
They called one full day a sleep, and a year was called a winter. Each month was referred to as a moon and always began with a new moon. An Indian day wasn’t divided into minutes or hours. It had four time periods—sunrise, noon, sunset, and midnight. Each season was determined by the budding or leafing of plants, the spawning of fish or the rutting time for animals. Most Indians thought the white race had been running around like scared rabbits ever since the invention of the clock.
The lexicon, or vocabulary, of a language is an inventory of the items a culture talks about and has categorized in order to make sense of the world and deal with it effectively. For example, modern life is dictated for many by the need to travel by some kind of vehicle—cars, trucks, SUVs, trains, buses, etc. We therefore have thousands of words to talk about them, including types of vehicles, models, brands, or parts.
The most important aspects of each culture are similarly reflected in the lexicon of its language. Among the societies living in the islands of Oceania in the Pacific, fish have great economic and cultural importance. This is reflected in the rich vocabulary that describes all aspects of the fish and the environments that islanders depend on for survival. For example, in Palau there are about 1,000 fish species and Palauan fishermen knew, long before biologists existed, details about the anatomy, behavior, growth patterns and habitat of most of them—in many cases far more than modern biologists know even today. Much of fish behavior is related to the tides and the phases of the moon. Throughout Oceania, the names given to certain days of the lunar months reflect the likelihood of successful fishing. For example, in the Caroline Islands, the name for the night before the new moon is otolol, which means “to swarm.” The name indicates that the best fishing days cluster around the new moon. In Hawai`i and Tahiti two sets of days have names containing the particle `ole or `ore; one occurs in the first quarter of the moon and the other in the third quarter. The same name is given to the prevailing wind during those phases. The words mean “nothing,” because those days were considered bad for fishing as well as planting.
Parts of Whorf’s hypothesis, known as linguistic relativity, were controversial from the beginning, and still are among some linguists. Yet Whorf’s ideas now form the basis for an entire sub-field of cultural anthropology: cognitive or psychological anthropology. A number of studies have been done that support Whorf’s ideas. Linguist George Lakoff’s work looks at the pervasive existence of metaphors in everyday speech that can be said to predispose a speaker’s world view and attitudes on a variety of human experiences. A metaphor is an expression in which one kind of thing is understood and experienced in terms of another entirely unrelated thing; the metaphors in a language can reveal aspects of the culture of its speakers. Take, for example, the concept of an argument. In logic and philosophy, an argument is a discussion involving differing points of view, or a debate. But the conceptual metaphor in American culture can be stated as ARGUMENT IS WAR. This metaphor is reflected in many expressions of the everyday language of American speakers: I won the argument. He shot down every point I made. They attacked every argument we made. Your point is right on target. I had a fight with my boyfriend last night. In other words, we use words appropriate for discussing war when we talk about arguments, which are certainly not real war. But we actually think of arguments as a verbal battle that often involve anger, and even violence, which then structures how we argue.
To illustrate that this concept of argument is not universal, Lakoff suggests imagining a culture where an argument is not something to be won or lost, with no strategies for attacking or defending, but rather as a dance where the dancers’ goal is to perform in an artful, pleasing way. No anger or violence would occur or even be relevant to speakers of this language, because the metaphor for that culture would be ARGUMENT IS DANCE.
Language and Identity
The way we speak can be seen as a marker of who we are and with whom we identify. We talk like the other people around us: where we live, our social class, our region of the country, our ethnicity, and even our gender. These categories are not homogeneous. All New Yorkers do not talk exactly the same; all women do not speak according to stereotypes; all African-Americans do not speak an African-American dialect. No one speaks the same way in all situations and contexts, but there are some consistencies in speaking styles that are associated with many of these categories.
As discussed above, people can indicate social class by the way they speak. The closer to the standard version their dialect is, the more they are seen as a member of a higher social class because the dialect reflects a higher level of education. In American culture, social class is defined primarily by income and net worth, and it is difficult (but not impossible) to acquire wealth without a high level of education. However, the speech of people in the higher social classes also varies with the region of the country where they live, because there is no single standard of American English, especially with respect to pronunciation. An educated Texan will sound different from an educated Bostonian, but they will use the standard version of English from their own region. The lower the social class of a community, the more their language variety will differ from both the standard and from the vernaculars of other regions.
An ethnicity, or ethnic group, is a group of people who identify with each other based on some combination of shared cultural heritage, ancestry, history, country of origin, language, or dialect. In the United States such groups are frequently referred to as “races,” but there is no such thing as biological race, and this misconception has historically led to racism and discrimination. Because of the social implications and biological inaccuracy of the term “race,” it is often more accurate and appropriate to use the terms ethnicity or ethnic group. A language variety is often associated with an ethnic group when its members use language as a marker of solidarity. They may also use it to distinguish themselves from a larger, sometimes oppressive, language group when they are a minority population.
A familiar example of an oppressed ethnic group with a distinctive dialect is African-Americans. They have a unique history among minorities in the United States, with their centuries-long experience as captive slaves and subsequent decades under Jim Crow laws. (These laws restricted their rights after their emancipation from slavery.) With the Civil Rights Acts of 1964 and 1968 and other laws, African-Americans gained legal rights to access public places and housing, but it is not possible to eliminate racism and discrimination only by passing laws; both still exist among the white majority. It is no longer culturally appropriate to openly express racism, but it is much less frowned upon to express negative attitudes about African-American Vernacular English (AAVE). Typically, it is not the language itself that these attitudes are targeting; it is the people who speak it.
As with any language variety, AAVE is a complex, rule-driven, grammatically consistent language variety, a dialect of American English with a distinctive history. A widely accepted hypothesis of the origins of AAVE is as follows. When Africans were captured and brought to the Americas, they brought their own languages with them. But some of them already spoke a version of English called a pidgin. A pidgin is a language that springs up out of a situation in which people who do not share a language must spend extended amounts of time together, usually in a working environment. Pidgins are the only exception to the Language Universal number 3 (all languages are systematic, rule driven, and equally complex overall, and equally capable of expressing any idea that the speaker wishes to convey).
There are no primitive languages, but a pidgin is a simplified language form, cobbled together based mainly on one core language, in this case English, using a small number of phonemes, simplified syntactic rules, and a minimal lexicon of words borrowed from the other languages involved. A pidgin has no native speakers; it is used primarily in the environment in which it was created. An English-based pidgin was used as a common language in many areas of West Africa by traders interacting with people of numerous language groups up and down the major rivers. Some of the captive Africans could speak this pidgin, and it spread among them after the slaves arrived in North America and were exposed daily to English speakers. Eventually, the use of the pidgin expanded to the point that it developed into the original forms of what has been called a Black English plantation creole. A creole is a language that develops from a pidgin when it becomes so widely used that children acquire it as one of their first languages. In this situation it becomes a more fully complex language consistent with Universal number 3.
All African-Americans do not speak AAVE, and people other than African-Americans also speak it. Anyone who grows up in an area where their friends speak it may be a speaker of AAVE like the rapper Eminem, a white man who grew up in an African-American neighborhood in Detroit. Present-day AAVE is not homogeneous; there are many regional and class variations. Most variations have several features in common, for instance, two phonological features: the dropped /r/ typical of some New York dialects, and the pronunciation of the “th” sound of words like this and that as a /d/ sound, dis and dat. Most of the features of AAVE are also present in many other English dialects, but those dialects are not as severely stigmatized as AAVE is. It is interesting, but not surprising, that AAVE and southern dialects of white English share many features. During the centuries of slavery in the south, African-American slaves outnumbered whites on most plantations. Which group do you think had the most influence on the other group’s speech? The African-American community itself is divided about the acceptability of AAVE. It is probably because of the historical oppression of African-Americans as a group that the dialect has survived to this day, in resistance to the majority white society’s disapproval.
In any culture that has differences in gender role expectations—and all cultures do—there are differences in how people talk based on their sex and gender identity. These differences have nothing to do with biology. Children are taught from birth how to behave appropriately as a male or a female in their culture, and different cultures have different standards of behavior. It must be noted that not all men and women in a society meet these standards, but when they do not they may pay a social price. Some societies are fairly tolerant of violations of their standards of gendered behavior, but others are less so.
In the United States, men are generally expected to speak in a low, rather monotone pitch; it is seen as masculine. If they do not sound sufficiently masculine, American men are likely to be negatively labeled as effeminate. Women, on the other hand, are freer to use their entire pitch range, which they often do when expressing emotion, especially excitement. When a woman is a television news announcer, she will modulate the pitch of her voice to a sound more typical of a man in order to be perceived as more credible. Women tend to use minimal responses in a conversation more than men. These are the vocal indications that one is listening to a speaker, such as m-hm, yeah, I see, wow, and so forth. They tend to face their conversation partners more and use more eye contact than men. This is one reason women often complain that men do not listen to them.
Deborah Tannen, a professor of linguistics at Georgetown University in Washington, D.C., has done research for many years on language and gender. Her basic finding is that in conversation women tend to use styles that are relatively cooperative, to emphasize an equal relationship, while men seem to talk in a more competitive way in order to establish their positions in a hierarchy. She emphasizes that both men and women may be cooperative and competitive in different ways.
Other societies have very different standards for gendered speech styles. In Madagascar, men use a very flowery style of talk, using proverbs, metaphors and riddles to indirectly make a point and to avoid direct confrontation. The women on the other hand speak bluntly and say directly what is on their minds. Both admire men’s speech and think of women’s speech as inferior. When a man wants to convey a negative message to someone, he will ask his wife to do it for him. In addition, women control the marketplaces where tourists bargain for prices because it is impossible to bargain with a man who will not speak directly. It is for this reason that Malagasy women are relatively independent economically.
In Japan, women were traditionally expected to be subservient to men and speak using a “feminine” style, appropriate for their position as wife and mother, but the Japanese culture has been changing in recent decades so more and more women are joining the work force and achieving positions of relative power. Such women must find ways of speaking to maintain their feminine identities and at the same time express their authority in interactions with men, a challenging balancing act. Women in the United States do as well, to a certain extent. Even Margaret Thatcher, prime minister of England, took speech therapy lessons to “feminize” her language use while maintaining an expression of authority.
The Deaf 'Culture' and Signed Languages
Deaf people constitute a linguistic minority in many societies worldwide based on their common experience of life. This often results in their identification with a local Deaf culture. Such a culture may include shared beliefs, attitudes, values, norms, and values, like any other culture, and it is invariably marked by communication through the use of a sign language. It is not enough to be physically deaf (spelled with a lower case “d”) to belong to a Deaf culture (written with a capital “D”). In fact, one does not even need to be deaf. Identification with a Deaf culture is a personal choice. It can include family members of deaf people or anyone else who associates with deaf people, as long as the community accepts them. Especially important, members of Deaf culture are expected to be competent communicators in the sign language of the culture. In fact, there have been profoundly deaf people who were not accepted into the local Deaf community because they could not sign. In some deaf schools, at least in the United States, the practice has been to teach deaf children how to lip read and speak orally, and to prevent them from using a signed system. They were expected to blend in with the hearing community as much as possible. This is called the oralist approach to education, but it is considered by members of the Deaf community to be a threat to the existence of their culture. For the same reason, the development of cochlear implants, which can restore hearing for some deaf children, has been controversial in U.S. Deaf communities. The members often have a positive attitude toward their deafness and do not consider it to be a disability. To them, regaining hearing represents disloyalty to the group and a desire to leave it.
According to the World Federation of the Deaf, there are over 200 distinct sign languages in the world, which are not mutually comprehensible. They are all considered by linguists to be true languages, consistent with linguistic definitions of all human languages. They differ only in the fact that they are based on a gestural-visual rather than a vocal-auditory sensory mode. Each is a true language with basic units comparable to phonemes but composed of hand positions, shapes, and movements, plus some facial expressions. Each has its own unique set of morphemes and grammatical rules. American Sign Language (ASL), too, is a true language separate from English; it is not English on the hands. Like all other signed languages, it is possible to sign with a word-for-word translation from English, using finger spelling for some words, which is helpful in teaching the deaf to read, but they prefer their own language, ASL, for ordinary interactions. Of course, Deaf culture identity intersects with other kinds of cultural identity, like nationality, ethnicity, gender, class, and sexual orientation, so each Deaf culture is not only small but very diverse.
Language Change
All languages change over time. In fact, it is not possible to keep them from doing so. How and why does this happen? The study of how languages change is known as historical linguistics. The processes, both historical and linguistic, that cause language change can affect all of its systems: phonological, morphological, lexical, syntactic, and semantic... The short answer is tweaks among and between the seven features of language above!
Historical linguists have placed most of the languages of the world into taxonomies, groups of languages classified together based on words that have the same or similar meanings. Language taxonomies create something like a family tree of languages. For example, words in the Romance family of languages, called sister languages, show great similarities to each other because they have all derived from the same “mother” language, Latin (the language of Rome). In turn, Latin is considered a “sister” language to Sanskrit (once spoken in India and now the mother language of many of India’s modern languages, and still the language of the Hindu religion) and classical Greek. Their “mother” language is called “Indo-European,” which is also the mother (or grandmother!) language of almost all the rest of European languages.
Let’s briefly examine the history of the English language as an example of these processes of change. England was originally populated by Celtic peoples, the ancestors of today’s Irish, Scots, and Welsh. The Romans invaded the islands in the first-century AD, bringing their Latin language with them. This was the edge of their empire; their presence there was not as strong as it was on the European mainland. When the Roman Empire was defeated in about 500 AD by Germanic speaking tribes from northern Europe (the “barbarians”), a number of those related Germanic languages came to be spoken in various parts of what would become England. These included the languages of the Angles and the Saxons, whose names form the origin of the term Anglo-Saxon and of the name of England itself—Angle-land. At this point, the languages spoken in England included those Germanic languages, which gradually merged as various dialects of English, with a small influence from the Celtic languages, some Latin from the Romans, and a large influence from Viking invaders. This form of English, generally referred to as Old English, lasted for about 500 years. In 1066 AD, England was invaded by William the Conqueror from Normandy, France. New French rulers brought the French language. French is a Latin-based language, and it is by far the greatest source of the Latin-based words in English today; almost 10,000 French words were adopted into the English of the time period. This was the beginning of Middle English, which lasted another 500 years or so.
The change to Modern English had two main causes. One was the invention of the printing press in the fifteenth century (and resulting boom of 'literacy' in Ong's terms), which resulted in a deliberate effort to standardize the various dialects of English, mostly in favor of the dialect spoken by the elite. The other source of change, during the fifteenth and sixteenth centuries, was a major shift in the pronunciation of many of the vowels (known, not terribly creatively, as The Great Vowel Shift). Middle English words like hus and ut came to be pronounced house and out. Many other vowel sounds also changed in a similar manner.
None of the early forms of English are easily recognizable as English to modern speakers. Here is an example of the first two lines of the Lord’s Prayer in Old English, from 995 AD, before the Norman Invasion:
Fæder ūre, ðū ðē eart on heofonum,
Sī ðīn nama gehālgod.
Here are the same two lines in Middle English, English spoken from 1066 AD until about 1500 AD. These are taken from the Wycliffe Bible in 1389 AD:
Our fadir that art in heuenes,
halwid be thi name.
The following late Middle English/early Modern English version from the 1526 AD Tyndale Bible, shows some of the results of grammarians’ efforts to standardize spelling and vocabulary for wider distribution of the printed word due to the invention of the printing press:
O oure father which arte in heven,
halowed be thy name.
And finally, this example is from the King James Version of the Bible, 1611 AD, in the early Modern English language of Shakespeare. It is almost the same archaic form that modern Christians use.
Our father which art in heauen,
hallowed be thy name.
Over the centuries since the beginning of Modern English, it has been further affected by exposure to other languages and dialects worldwide. This exposure brought about new words and changed meanings of old words. More changes to the sound systems resulted from phonological processes that may or may not be attributable to the influence of other languages. Many other changes, especially in recent decades, have been brought about by cultural and technological changes that require new vocabulary to deal with them.
Consider the following...
Try This: Just think of all the words we use today that have either changed their primary meanings, or are completely new: mouse and mouse pad, Google, app, computer (which used to be a person who computes!), text, phone, etc. How many more can you think of?
Globalization and Language Change
Globalization is the spread of people, their cultures and languages, products, money, ideas, and information around the world. Globalization is nothing new; it has been happening throughout the existence of humans, but for the last 500 years it has been increasing in its scope and pace, primarily due to improvements in transportation and communication. Beginning in the fifteenth-century, English explorers started spreading their language to colonies in all parts of the world. English is now one of the three or four most widely spoken languages. It has official status in at least 60 countries, and it is widely spoken in many others. Other colonizers also spread their languages, especially Spanish, French, Portuguese, Arabic, and Russian. Like English, each has its regional variants. One effect of colonization has often been the suppression of local languages in favor of the language of the more powerful colonizers.
In the past half century, globalization has been dominated by the spread of North American popular culture and language to other countries. Today it is difficult to find a country that does not have American music, movies and television programs, or Coca Cola and McDonald’s, or many other artifacts of life in the United States, and the English terms that go with them.
In addition, people are moving from rural areas to cities in their own countries, or they are migrating to other countries in unprecedented numbers. Many have moved because they are refugees fleeing violence, or they found it increasingly difficult to survive economically in their own countries. This mass movement of people has led to the on-going extinction of large numbers of the world’s languages as people abandon their home regions and language in order to assimilate into their new homes.
Language Shift, Language Maintenance, and Language Death
Of the approximately 6,000-7,000 languages still surviving today, about half the world’s more than seven billion people speak only ten. These include Mandarin Chinese, two languages from India, Spanish, English, Arabic, Portuguese, Russian, Japanese, and German. Many of the rest of the world’s languages are spoken by a few thousand people, or even just a few hundred, and most of them are threatened with extinction, called language death. It has been predicted that by the end of this century up to 90 percent of the languages spoken today will be gone. The rapid disappearance of so many languages is of great concern to linguists and anthropologists alike. When a language is lost, its associated culture and unique set of knowledge and worldview are lost with it forever. Remember Whorf’s hypothesis.
Some minority languages are not threatened with extinction, even those that are spoken by a relatively small number of people. Others, spoken by many thousands, may be doomed. What determines which survive and which do not? Smaller languages that are associated with a specific country are likely to survive. Others that are spoken across many national boundaries are also less threatened, such as Quechua, an indigenous language spoken throughout much of South America, including Colombia, Ecuador, Peru, Chile, Bolivia, and Argentina. The great majority of the world’s languages are spoken by people with minority status in their countries. After all, there are only about 193 countries in the world, and over 6,000 languages are spoken in them. You can do the math.
The survival of the language of a given speech community is ultimately based on the accumulation of individual decisions by its speakers to continue using it or to abandon it. The abandonment of a language in favor of a new one is called language shift. These decisions are usually influenced by the society’s prevailing attitudes. In the case of a minority speech community that is surrounded by a more powerful majority, an individual might keep or abandon the native language depending on a complex array of factors. The most important factors will be the attitudes of the minority people toward themselves and their language, and the attitude of the majority toward the minority.
Language represents a marker of identity, an emblem of group membership and solidarity, but that marker may have a downside as well. If the majority look down on the minority as inferior in some way and discriminates against them, some members of the minority group may internalize that attitude and try to blend in with the majority by adopting the majority’s culture and language. Others might more highly value their identity as a member of that stigmatized group, in spite of the discrimination by the majority, and continue to speak their language as a symbol of resistance against the more powerful group. One language that is a minority language when spoken in the United States and that shows no sign of dying out either there or in the world at large, is Spanish. It is the primary language in many countries and in the United States it is by far the largest minority language.
Another example of an oppressed minority group that has struggled with language and culture loss is Native Americans. Many were completely wiped out by the European colonizers, some by deliberate genocide but the great majority (up to 90 percent) by the diseases that the white explorers brought with them, against which the Native Americans had no immunity. In the twentieth-century, the American government stopped trying to kill Native Americans but instead tried to assimilate them into the white majority culture. It did this in part by forcing Native American children to go to boarding schools where they were required to cut their hair, practice Christianity, and speak only English. When they were allowed to go back home years later, they had lost their languages and their culture, but had not become culturally “white” either. The status of Native Americans in the nineteenth and twentieth-centuries as a scorned minority prompted many to hide their ethnic identities even from their own children. In this way, the many hundreds of original Native American languages in the United States have dwindled to less than 140 spoken today, according to UNESCO. More than half of those could disappear in the next few years, since many are spoken by only a handful of older members of their tribes. However, a number of Native American tribes have recently been making efforts to revive their languages and cultures, with the help of linguists and often by using texts and old recordings made by early linguists like Edward Sapir.
How is the Digital Age Changing Communication?
The invention of the printing press in the fifteenth-century was just the beginning of technological transformations that made the spread of information in European languages and ideas possible across time and space using the printed word. Recent advances in travel and digital technology are rapidly transforming communication; now we can be in contact with almost anyone, anywhere, in seconds. However, it could be said that the new age of instantaneous access to everything and everyone is actually continuing a social divide that started with the printing press.
In the fifteenth-century, few people could read and write, so only the tiny educated minority were in a position to benefit from printing. Today, only those who have computers and the skills to use them, the educated and relatively wealthy, have access to this brave new world of communication. Some schools have adopted computers and tablets for their students, but these schools are more often found in wealthier neighborhoods. Thus, technology is continuing to contribute to the growing gap between the economic haves and the have-nots.
There is also a digital generation gap between the young, who have grown up with computers, and the older generations, who have had to learn to use computers as adults. These two generations have been referred to as digital natives and digital immigrants. The difference between the two groups can be compared to that of children versus adults learning a new language; learning is accomplished much more easily by the young.
Computers, and especially social media, have made it possible for millions of people to connect with each other for purposes of political activism, including “Occupy Wall Street” in the United States and the “Arab Spring” in the Middle East. Some anthropologists have introduced computers and cell phones to the people they studied in remote areas, and in this way they were able to stay in contact after finishing their ethnographic work. Those people, in turn, were now able to have greater access to the outside world.
Facebook and 'micro-blogging' platforms (like X/Twitter, Bluesky, etc.) are becoming key elements in the survival of a number of endangered indigenous languages. Facebook is now available in over 70 languages, and Twitter in about 40 languages. For example, a website has been created that seeks to preserve Anishinaabemowin, an endangered Native American language from Michigan.
The language has 8,000-10,000 speakers, but most of the native speakers are over 70 years old, which means the language is threatened with extinction. Modern social media are an ideal medium to help encourage young people to communicate in their language to keep it alive.
Clearly, language and communication through modern technology are in the forefront of a rapidly changing world, for better or for worse. It’s anybody’s guess what will happen next.
Language Change over Generations
While the notion of discrete "generations" as bounded demographic categories—Baby Boomers, Generation X, Millennials, Generation Z—itself represents a largely culturally fabricated construct rather than a biological or linguistic reality, the phenomenon of age-graded language variation remains fundamentally important to understanding language change. Generations, however we choose to delineate them, do indeed produce language change, but they do so primarily because they produce identity change. Young speakers do not merely inherit the linguistic systems of their predecessors whole cloth; rather, they actively or accidentally reshape it as part of constructing their own social identities distinct from those of their parents and other authority figures. This process operates at multiple levels simultaneously: young speakers adopt novel vocabulary, innovate grammatical structures, shift phonological patterns, and repurpose existing linguistic resources to signal membership in their peer groups while simultaneously marking distance from older generations.
You may be able to think of goofy things you say in your family simply because a child mispronounced them, but generational langauge change goes well beyond this phenomenon.
The Sapir-Whorf hypothesis offers a useful way for understanding this type of language change: if language shapes thought and perception, then different generations, speaking subtly (or not so subtly) different varieties of what we call "the same language," quite literally inhabit different worlds. A teenager in 2025 who describes something as "bussin" or uses "slay" as a verb of approval is not simply using different words for the same concepts that their grandparents might express as "excellent" or "wonderful" or their grandparents may have called "capital"—they are participating in an entirely different semiotic system, one that indexes different cultural values, different media ecosystems, different social hierarchies, and different ways of constructing the self in relation to others.
This divergence in linguistic practice inevitably generates what we might term generational moral panic. A moral panic is a period of broad fear and anxiety where a group or cultural behavior is falsely or exaggeratedly portrayed as a threat to society's values. Moral panics are a recurring pattern in which older generations perceive the innovations (including linguistic changes) of younger speakers not simply as different, but as evidence of cognitive decline, cultural degradation, or moral decay. Interestingly, moral panics often correspond with the introduction of new media forms. Broad moral panic surrounded the advent of the printing press, newspapers, radio, television, Dungeons & Dragons, video games, the internet, video games (again), and--most recently--various forms of artificial intelligence.
Such moral panics about language are remarkably consistent across time periods and cultures, suggesting that they serve important social functions in negotiating intergenerational power dynamics and cultural authority. In contemporary America, we can observe this phenomenon vividly in middle school contexts, where linguistic innovations spread with particular rapidity through peer networks. Consider the emergence of the "six seven" verbal expression and hand gesture among middle schoolers in certain regions and online communities in 2025. Originating from the song "Doot Doot (6 7)" by Skrilla, the expression took on meme status in part thanks to its use in TikTok edits. While the phrase lacks a fixed meaning, other (sometimes coincidental) usages have come to serve other localized meanings depending on context. In any event, "six seven" is best understood by linguistic anthropologists as 'sticky' or attractive to young speakers because of a simple feature: it exemplifies how young speakers create and circulate linguistic forms that are deliberately opaque to adults. The very inscrutability of such expressions to parents and teachers constitutes part of their social value. They function as identity markers that simultaneously build in-group solidarity and exclude outsiders. Predictably, adult reactions to such linguistic innovations often frame them as evidence that "kids these days can't communicate properly" or that "[texting / TikTok / YouTube / the internet / video games / etc.] is destroying literacy"—moral judgments that conflate linguistic difference with linguistic deficiency.
Yet these contemporary anxieties are hardly new. They echo concerns that have attended language change across centuries. In early twentieth-century America, social commentators lamented the "slovenly speech" of young people who had adopted jazz-age slang like "the bee's knees" or "the cat's pajamas"—expressions that today seem quaintly innocent but which at the time were perceived by many adults as vulgar corruptions of proper English that presaged the collapse of civilized discourse. Similarly, the emergence of youth-oriented slang in the 1960s ("groovy," "far out," "dig it") was frequently interpreted by older Americans not merely as generational difference but as symptomatic of broader social unraveling associated with the counterculture movement. The linguistic innovations were inseparable from—and indeed, constitutive of—the identity transformations that characterized that era.
Looking beyond American contexts, we find remarkably similar patterns across vastly different historical periods. Ancient Roman writers, including Cicero and Quintilian, regularly complained about the declining standards of Latin among young people, who were accused of adopting vulgar colloquialisms, foreign borrowings, and simplified grammatical structures that deviated from classical norms. What these Roman critics were actually observing, of course, was language change in progress—the gradual evolution of Classical Latin toward the various Romance languages that would eventually emerge. Their moral panic about linguistic corruption was simultaneously a panic about social change, political transformation, and the apparent erosion of traditional Roman values.
The Elizabethan / Shakespearean era provides another instructive example of the complex relationship between generational language change and moral anxiety. Shakespeare himself was a prodigious creator and popularizer of neologisms, introducing or giving currency to thousands of words and phrases that enriched English but which were often criticized by contemporary purists as unnecessary corruptions of the language. Words like "addiction," "bedroom," "eyeball," and "fashionable"—now thoroughly unremarkable—were once viewed by some of Shakespeare's contemporaries as the kind of flashy, youth-oriented linguistic innovation that degraded the dignity of English. But linguistic historians argue about whether Shakespeare really 'invented' all of these words or was simply the first to put the words he was hearing into writing. So to say, Shakespeare's linguistic innovations were not merely individual creativity—they reflected and accelerated broader changes in Early Modern English driven partly by generational turnover in a period of rapid social change, expanded literacy, and increased contact with other languages.
The anxiety about these changes was palpable in works like Thomas Wilson's The Arte of Rhetorique (1553), which decried young scholars who "pouder their talke with ouersea language" and employ fancy foreign-derived vocabulary to show off, rather than communicating clearly in good plain English. What Wilson perceived as a moral failing was actually a generation of speakers navigating new social realities—increased international commerce, humanist education, religious reformation—that required new linguistic resources. The parallel to contemporary concerns about young people's language is striking: in both cases, linguistic innovation that serves genuine communicative and identity needs is reinterpreted through a lens of moral anxiety as evidence of decline. The "6 7" of middle school culture, the slang of the jazz age, the vulgarisms of young Romans, and the neologisms of Elizabethan youth all share a common feature—they represent generations using language not merely to communicate information, but to construct new forms of social identity that distinguish them from their predecessors, thereby ensuring that language, like the societies it serves, remains perpetually in motion.
The recurring correlation between moral panics and the emergence of new media forms offers insight into the relationship between language, culture, and generational anxiety. From wax writing tablets that horrified the ancient Greek philosopher Socrates (he thought we'd lose good human memory if we could write things donw), to the printing press that horrified medieval scribes, to the social media platforms that alarm contemporary parents—each new medium does not merely provide a novel channel for transmitting existing language. Rather, it fundamentally restructures the conditions under which language is produced, circulated, and interpreted, thereby reshaping the cultural worlds that speakers inhabit. In Sapir-Whorf terms, a generation raised with TikTok, Discord, and Snapchat does not simply communicate the same ideas through different technologies—they think, perceive, and construct reality differently than a generation whose linguistic socialization occurred primarily through face-to-face conversation, telephone calls, and handwritten letters as well as one somewhere in between who prefer Facebook (or perhaps remember Myspace).
When parents express alarm that their children communicate through memes, abbreviations like "6 7," or AI-generated text, they are responding not merely to surface-level linguistic differences but to a profound recognition that these young people are cognitively and culturally inhabiting a different world, one whose rules and values are not immediately transparent to those formed by earlier media ecologies. The moral panic, then, serves as a kind of social alarm system signaling that a genuine transformation in human consciousness and communication is underway—though the panic typically misdirects our attention toward imagined harms (literacy decline, shortened attention spans, moral corruption) rather than toward the actual adaptive challenge of understanding and navigating the new cultural-linguistic landscape being created. Linguistic anthropologists have an opportunity to study theses alarms!
History suggests that each generation ultimately accommodates itself to the media innovations of its youth, at which point those once-threatening technologies become naturalized as simply "the way things are"—until the next innovation arrives to provoke the next moral panic, continuing the cycle through which language, culture, and human perception evolve together across generational time. So what's next?
Modern Language and Large Language Models (LLM, "AI")
The discussion of generational moral panics and new media forms brings us inevitably to one of the most significant technological developments of our current moment: Large Language Models and artificial intelligence. It is worth stating explicitly at the outset that what follows is not itself a moral panic about AI, nor an argument against its use. Just as previous generations had to grapple thoughtfully with the genuine transformations brought about by the printing press, television, or the internet—beyond the initial alarm and anxiety—our generation faces the task of understanding AI's relationship to language not through fear or uncritical embrace, but through careful examination of what changes when machines enter the fundamentally human domain of meaning-making. The goal here is not to condemn or celebrate, but to think clearly about what is at stake when we invite AI into our most fundamental cognitive and social practice in the uses of language itself.
The emergence of Large Language Models (LLMs)—artificial intelligence (AI) systems trained on vast corpora (the plural of 'corpus' to make Cicero and Quintilian above happy) of human-written text to generate statistically probable sequences of words—presents linguistic science with a curious paradox. These systems produce output that superficially resembles human language so convincingly that we are predisposed to accept it as language, yet what LLMs do fundamentally differs from what humans do when we use language. An LLM does not possess intentions, experiences, or a model of the world. It does not mean anything when it produces text. Rather, it performs extraordinarily sophisticated pattern-matching, predicting which word is most likely to follow based on the statistical regularities it has learned from billions or trillions of examples of human writing.
When you read a sentence generated by an LLM, your brain—evolved over millennia to interpret linguistic signals as emanating from other conscious beings with communicative intentions—automatically supplies the meaning, the intentionality, the human presence that the text itself lacks. (This social-psychological relationship is further complicated by our 'prompting' the AI to produce a certain expected result.) The output meets our expectations for language on its surface. It has grammar, coherence, even apparent logic and personality. This surface-level adequacy is precisely what makes LLMs so powerful and, simultaneously, so potentially problematic for our understanding of what language actually is. We see sentences, paragraphs, even entire essays that look like language, sound like language, and function like language in many contexts—and our cognitive systems, not evolved to distinguish statistically-generated text from intentionally communicative utterances, treat them as equivalent.
More interesting still is the feedback loop that LLMs create within the broader linguistic ecology. The training data for these models consists primarily of human-written text scraped from the internet, books, articles, and other sources—a vast repository of genuine human communication reflecting diverse voices, styles, and purposes. The models learn the patterns in this data and generate new text based on those patterns. Increasingly, however, that generated text itself circulates back into the world. Students use AI to write essays, journalists use it to draft articles, businesses use it to generate marketing copy, social media users employ it to craft posts and responses, and so on. This AI-generated text then becomes part of the linguistic environment in which humans read, write, and learn—and inevitably, some of it gets incorporated into the training data for the next generation of LLMs. We are thus creating a recursive system in which human language trains AI, AI generates synthetic language, humans are exposed to and potentially influenced by that synthetic language (consciously or unconsciously adopting its patterns, phrasings, and structures), and this hybrid linguistic product then trains future AI systems... The long-term consequences of this feedback loop remain unclear, but the potential for what we might call linguistic drift—a gradual homogenization or flattening of language as distinctively human voices are "averaged out" through repeated AI mediation—deserves serious consideration.
Consider, then, how AI may already be impacting the way you speak, write, and interact with fellow humans. Have you found yourself using phrases or constructions that "sound like" AI-generated text? When drafting an email or essay, do you sometimes mentally preview how an AI might phrase something, or feel your own writing converging toward the smooth, inoffensive, sycophantic/pandering, generically competent style that characterizes much LLM output? Do you find yourself interacting with other humans through templates and patterns rather than through genuinely spontaneous, contextually responsive communication? These are not purely rhetorical questions but genuine prompts for self-reflection. If, as the Sapir-Whorf hypothesis suggests, our language shapes how we perceive and conceptualize reality, then changes in our linguistic practices—including those induced by regular interaction with and reliance upon AI-generated text—may be changing not just how we communicate but how we think and what we are capable of thinking.
The question facing our generation, then, is fundamentally about cognitive and cultural autonomy: what aspects of language, and therefore of thought and worldview, can we safely delegate to AI systems? Where should we insist on preserving distinctively human linguistic practices—messy, inefficient, emotionally charged, and unpredictable as they may be—as essential to maintaining our humanity? The convenience of having AI compose our emails, draft our reports, or even engage in our social media conversations must be weighed against the possibility that in outsourcing language production, we may be outsourcing something more fundamental: the uniquely human capacity to mean something, to intend something, to reach across the unbridgeable gap between consciousnesses and touch another mind with our words.
Language is not merely a tool for transmitting information. LLMs are good enough at that. Look again at the seven features of language that opened this chapter. While LLMs can simulate several of these features, they fundamentally lack the grounding that makes them genuinely linguistic. Cultural transmission requires embodied experience within a community—LLMs are trained on decontextualized text, not (yet) raised in social interaction. Arbitrariness and productivity appear intact until we recognize that LLMs don't understand words as arbitrary symbols agreed upon by a community, nor do they intentionally repurpose language based on shifting cultural needs (they're still bad at developing new jokes, for example). Most critically, they lack pragmatic competence. Without genuine understanding of social context, speaker relationships, or communicative intent, they are pattern-matching rather than meaning-making—a fundamental distinction that separates statistical approximation from actual language use. Language is the medium through which we construct our shared reality, negotiate our relationships, and ultimately constitute ourselves as social beings. The question is not whether AI will become part of our linguistic landscape—it already is—but rather how we will navigate that landscape in ways that preserve what is essential about human communication while adapting to the new realities of a world in which not all language comes from human minds.
Language as Living Practice
Language, as we have seen throughout these chapters, is far more than a static system of symbols or a mere tool for transmitting information. It is a dynamic, living practice through which we construct our identities, navigate our social worlds, and ultimately shape our perception of reality itself. From the fundamental building blocks of phonemes and morphemes to the complex interplay of syntax and semantics, from the ways children acquire language to the mechanisms through which it changes across generations, every aspect of linguistic study reveals the same essential truth: language and humanity are inseparable.
We are not beings who happen to use language; we are beings constituted through language, our very capacity for thought and culture inextricably bound to the symbolic systems we inherit, transform, and pass on.
The challenges facing language in our contemporary moment—the integration of artificial intelligence into our communicative practices, the rapid evolution of digital media, the perpetual tension between innovation and tradition—are not aberrations but continuations of processes as old as language itself. Every generation has faced the task of adapting language to new realities while preserving the essential human capacities that make communication possible. What remains constant is the need for critical awareness and understanding not just how language works, but what it does, what it means, and what is at stake when we choose—or fail to choose—how we will speak, write, and mean in an ever-changing world.
Language is power, identity, thought, and culture all at once. How humans wield it, preserve it, and allow it to evolve will determine not just how we communicate, but who we are and what we can become.
