There is between the whole animal kingdom on the one side, and man, even in his lowest state, on the other, a barrier which no animal has ever crossed, and that barrier is—language. — Friedrich Max Müller1

One of the most basic teachings of the Christian faith is that humans are the only creatures on earth made in the image of God (Gen. 1:26-28). We are God’s vice-regents, given the special privilege of ruling over creation on his behalf.2 However, with the rise of Darwinian naturalism, this supposedly “anthropocentric” view of the universe has been increasingly challenged. Humans are said to be nothing more than advanced apes, differing from the rest of the animal kingdom only in degree, not in kind. We are told that the forces of random mutation and natural selection can account for every aspect of our existence, without recourse to any “god of the gaps.”

But there is at least one “gap” that stubbornly refuses to go away: language. Even secular scientists admit that there is something about human language that sets us apart from the animals. But what exactly is language? What makes it different from other types of communication? And what does it have to do with the Christian doctrine of the imago Dei?

Defining Language

One of the earliest attempts at scientifically defining language goes back to a linguist named Charles Hockett.3 In the 1950s, Hockett created a list known as the “Design Features of Language”—features that, taken together, are found only among us humans. Hockett’s list included:

Semanticity. Most animals are only capable of expressing their own internal state. Their growls, chirps, and coos simply mean that they feel threatened, or hungry, or willing to mate. But semantic communication has reference to something outside the organism itself, such as the presence of a predator or food source. This sometimes makes it difficult to define precisely when animal communication counts as truly semantic. When a monkey screams at the sight of a cheetah, does he simply mean, “I’m really scared” (internal state), or does he mean, “There’s a cheetah over there” (semantic reference)? Whatever the case may be, animals are quite limited in their ability to “talk” about things beyond their internal state.

Displacement. Semantic limitations notwithstanding, some animals do seem capable of communicating basic things about their immediate surroundings. For example, prairie dogs have a call equivalent to, “I see a hawk,” which differs from their calls for land predators like coyotes. But no prairie dog seems capable of saying, “Hey, do you guys remember that hawk by the river last week?” Language therefore involves the ability to communicate about things displaced in time and space.

Arbitrariness. When animals communicate, there is almost always a fixed “iconic” connection between their signals and what they signify. For example, the meaning of a dog’s growl is pretty straightforward: “Back off!” But there is no inherent connection between the English word dog and the four-legged canine species itself. That is why the word can vary depending on the language (perro in Spanish, kelb in Arabic, and so on). The connection for humans is completely arbitrary. Only in rare settings (such as a gorilla being artificially taught sign language) does arbitrariness occur among animals.

Discreteness. In human language, every complete thought (sentence) can be broken down into individually meaningful segments (words). These individual units can then be recombined to form different meanings (“The cat chased the mouse” is quite different from “The mouse chased the cat”). This is why linguists refer to language as a discrete combinatorial system. By contrast, many forms of animal communication are “continuous.” That is, their communication patterns exist on a sliding scale. Honeybee dances would be a good example of continuous communication. By performing a unique dance on the surface of a honeycomb, bees can communicate the direction, distance, and quantity of a particular food source. But each aspect of the dance (such as the intensity of vibration) has a value such that between any two given values, an endless number of intermediate values is hypothetically possible. If words are like on/off switches, continuous communication is more like the turn of a dial.

Traditional transmission. Human languages do not come hardwired in our brains. They have to be learned, which is why they differ from culture to culture. By contrast, a honeybee dance is completely instinctive; no one has to teach the bee how to dance. Most forms of animal communication are innate (with few exceptions, such as the songs of certain songbirds). Wherever human language came from, it cannot simply be an evolutionary development on hardwired instincts.

Openness and productivity. This is one of the most remarkable features that sets human language apart from all animal communication. Animals are quite limited in the things they can talk about. Generally these things do not extend beyond immediate needs like feeding and mating. By contrast, human language is generative and unbounded—theoretically infinite in its capacity for novel expression.

The problem with such diagnostic lists is that they are subject to constant revision, and they can sometimes give the impression of setting an arbitrary standard. Sue Savage-Rumbaugh, a famous zoologist who has worked extensively teaching bonobos sign language, complains about linguists:

They keep raising the bar. First the linguist said we had to get our animals to use signs in a symbolic way if we wanted to say they learned language. OK, we did that, and then they said, “No, that’s not language, because you don’t have syntax.” So we proved our apes could produce combinations of signs, but the linguists said that wasn’t enough syntax, or the right syntax. They’ll never agree that we’ve done enough.4

However, it is not as though linguists are just trying to set the animals up for failure. There is more to the story than simple “anthropocentric” prejudice. As studies have progressed since the time of Hockett, linguists have learned more about the central role of syntax in making language what it is. In particular, the syntactic feature of recursion seems to be fundamental to the unbounded generative capacity of human language.


“I am your father’s brother’s nephew’s cousin’s former roommate!” “I want you to tell him all the information you just told me. I want him to know what I know. I want him to know I want him to know!”5 These are examples of recursion, the phenomenon of embedding linguistic units within other linguistic units of the same kind (think Russian nesting dolls). It includes such grammatical principles as possessive pronouns, relative clauses, and much more. This feature of language—so basic to our everyday speech—allows us to form a limitless number of novel expressions. It also allows us to think about our thinking and to talk about our talking.

Recursion is thus a key aspect of what Thomas Suddendorf calls “meta-representation,” a necessary ingredient for interpersonal conversation.6 In order to be mutually understood, our own thoughts must be represented in the minds of others, and vice versa. It also requires the ability to “take a step back” and think about the relationship between the symbols (words) we use and their referents. After all, if humans could never agree upon the meaning of any given symbol, we could never arrive at a common language. But that requires the ability to think about a symbol’s role as a symbol.

Recursion has received special attention in the writings of linguist Noam Chomsky. He has argued that our unique capacities for recursive thought must be hardwired in our brains, as a function of what he has called “Universal Grammar.”7 Languages themselves need to be learned, but the capacity for learning them must itself be innate. Therefore, Chomsky argued, animals can never be trained to achieve anything approximating genuine language. They simply lack the necessary hardware.8

Language as Image-Bearing

So what does language have to do with the Christian doctrine of the imago Dei? Quite a lot, as it turns out. Scripture never provides an explicit definition of the image of God, but it does provide a number of contextual clues. In Genesis 1-2, we read that humans are created in the context of a covenant. Broadly defined, a covenant is a solemn relationship between two parties, with mutual promises and obligations. Other elements of a covenant often include a historical prologue and threats for covenantal breach. In all of these respects, a covenant relationship requires language. Language is what enables us to recount history, make commands, offer promises, issue threats, and so forth. Further, the ability to reflect on the covenant relationship itself requires the capacity for recursive thought. If image-bearing implies a covenant relationship, and if a covenant relationship requires language, then we must conclude that language is an essential part of our identity as human beings.

Another contextual clue is the character of God himself. In Genesis 1, we read that God speaks the world into existence through his creative word. Language is thus essential not only to humanity, but also to divinity, which supports the idea that imaging God also means imitating God. Language enables us to act as “sub-creators,” reflecting the expressive creativity of God himself.

That language is a uniquely human trait should come as no surprise to Christians. We confess that God’s image is not a product of the blind, irrational forces of evolution, but rather a supernatural gift that has been divinely bestowed upon us. On this score, the scientific evidence weighs in favor of the Christian worldview. Humanity’s privileged place in the cosmos remains secure.

  1. Quoted in Thomas Suddendorf, The Gap: The Science of What Separates Us from Other Animals (Basic Books, 2013), 78.
  2. Theologians have debated exactly what the image of God entails. Three views have become popular: the resemblance (ontological) view says that God’s image is about possessing certain capacities; the representative (functional) view says that it is about exercising certain responsibilities; and the relational view says that it is about living in community with other people. However, these views do not have to be mutually exclusive. See C. John Collins, Genesis 1-4: A Linguistic, Literary, and Theological Commentary (P&R, 2005), 61-67.
  3. Stephen R. Anderson, Doctor Dolittle’s Delusion: Animals and the Uniqueness of Human Language (Yale University Press, 2004), 20-34.
  4. Ibid, 35.
  5. These quotes are taken from the films Spaceballs and Kill Bill: Vol. 1.
  6. Noam Chomsky, On Language (New Press, 1998).
  7. Suddendorf, 67.
  8. Chomsky’s theory of Universal Grammar has met with strong resistance on several fronts. On the one hand, some animal behaviorists have claimed that recursion can be found among certain animals, such as chickadees and starlings. On the other hand, some anthropologists claim to have found certain tribes that speak non-recursive languages, such as the Pirahã of Brazil. However, these contrary claims have yet to be confirmed, and most linguists remain skeptical that they will overturn Chomsky’s theory. See Con Slobodchikoff, Chasing Doctor Dolittle: Learning the Language of Animals (St. Martin’s Press, 2012), 23; Daniel Everett, “Cultural Constraints on Grammar and Cognition in Pirahã: Another Look at the Design Features of Human Language,” Current Anthropology 4 (Aug-Oct 2005): 621-646; Michael Corballis, The Recursive Mind: The Origins of Human Language, Thought, an Civilization (Princeton University Press, 2014), 25ff.