Thomas Aquinas was a big believer in the importance of distinguishing between necessary and sufficient conditions.
If you want to know how language began, you had better have some idea of what distinguishes language from all the other behaviors that can be found in the animal kingdom. On this blog, we are looking at how people gained the ability to tell one another about whatever random topic interests them. If I were a disciple of Noam Chomsky, I would be looking into the rise of recursive thinking. A pragmatics scholar would want to know about the ability to share an informative/communicative intention. Any other candidates?
In 2013 a paper by Kimbrough Oller et al. proposed, “Functional flexibility [the ability to express a range of emotions with the same sound] is a sine qua non in spoken language, because all words or sentences can be produced as expressions of varying emotional states and because learning conventional “meanings” requires the ability to produce sounds that are free of any predetermined function. Functional flexibility is a defining characteristic of language, and empirically it appears before syntax, word learning, and even earlier-developing features presumed to be critical to language (e.g., joint attention, syllable imitation, and canonical babbling).” This is a tricky one because a sine qua non is not the same thing as a distinguishing characteristic. For example, awareness of the world around one is a sine qua non of language, but it is not a distinguishing trait. You must be conscious to learn language, but lots of non-speaking animals are aware of the world around them. As the philosophers like to say, it is a necessary but not a sufficient condition.
Most insufficient conditions for language are widely shared. Awareness, the ability to focus attention, ability to associate signal with thing, etc. are all shared with many other species, so we don’t say they are precursors of language. If I returned from Borneo with news of a new ape, you would be interested. If I added that they were able to focus their attention and so were on the way to speaking, you would scoff. Functional flexibility, however, was a scientific unknown in non-humans and so Oller and the others commented prudently, “The appearance of functional flexibility early in the first year of human life is a critical step in the development of vocal language and may have been a critical step in the evolution of human language.”
I have argued on this blog and in my book that the rise of language was preceded by a very long (million+ years) period of emotional vocalizing. This vocalization was necessary to create a talking lineage, but it was not of itself sufficient to say that language itself existed during that period. Neither was language the inevitable outcome of this vocalization, which evolved for reasons of its own. Language then evolved in the face of new conditions.
So now, tada, an online paper by Zanna Clay et al announces, “We found that wild bonobos use a specific call type (the “peep”) across a range of contexts that cover the full valence range (positive-neutral-negative) in much of their daily activities, including feeding, travel, rest, aggression, alarm, nesting and grooming.”
Bonobos, chimpanzees and humans share a last common ancestor of about 6 million years ago. The last common bonobo and chimpanzee ancestor was probably less than 3 million years ago. As far as I know, functional flexibility is not reported in chimpanzees (although another look at the data might not be a bad idea). So let’s stick with what we know today. Two of three species descended from a common ancestor share a trait. Two explanations seem possible (1) the trait was found in the common ancestor and lost in the chimpanzee lineage, or (2) the trait was not found in the common ancestor and evolved separately in the human and bonobo lines.
Without knowing the genetic basis of the functional flexibility, it seems difficult to say which explanation is correct. I lean toward the second, partly because it agrees more with my past writing and partly because I suspect the trait probably evolved separately in dogs. It is common for dog yelps to sound alike to most people while dog owners and lovers can distinguish meanings between yelps. I would be surprised, frankly if elephants don’t show some functional flexibility in their sounds. So functional flexibility in vocalizations may be one of those many traits that pop up in a variety of circumstances.
For all the paper’s value, however, it has gotten my goat. My general complaint is the tendency of the authors (and they are hardly alone in this) to deny that language is a remarkable, even unique achievement. The closest thing to our ability to pay joint attention to topics is the dance of the honeybee, in which one member of a hive tells other members how to find pollen. It seems unlikely that our languages build on bee capacities, so an evolutionary account of language origins must explain where something new came from.
The authors of the bonobo paper do not further this task when they state: “natural selection acts by modifying or adding complexity to existing structures and mechanisms rather than by generating entirely new ones, a logic that has also been applied to the evolution of language.” First, let’s not forget that natural selection can simplify, and even lose, structures as well as complicate them. Second, new structures are unusual, but they are not unknown. One familiar example is the difference between bacteria and eukaryotes (every living thing that is not bacteria). Eukaryotes, by definition, have a cell nucleus. Bacteria don’t. That nucleus is a structure, selected no doubt by a natural process. When bacteria became eukaryotes natural selection gave us a new structure. My point is this: don’t be too quick to insist that you know how natural selection must work, especially when it is producing something not previously found in the tree of life.
The authors then go on to draw an entirely wrong conclusion from their insistence that they know what natural selection must have done: “A research goal therefore is to describe the basic design principles of early stages of vocal behaviour, which may have served as building blocks on which subsequent stages of linguistic development have emerged.” That process might be sufficient to show how the forefoot of a land mammal became a seal’s flipper, but what are the building blocks of topic selection, recursion, or informative/communicative intention?
Chomsky has been arguing for decades that you either have recursion or you don’t. I say the same thing about topic selection. Thomas Scott-Phillips argues the same way about informative/communicative intentions. You can point to the rise of necessary but insufficient conditions all you want, but at some time you have to face the unprecedented trait and say either it’s a mystery, or here is a possible solution.
The authors then say: “The aim of the current study is therefore to systematically analyse the acoustic structure of peeps to assess whether they are tied to specific behavioural contexts or whether, like human infants bonobos are capable of producing the same vocalisation across a range of affective states.” The aim is laudable and well worth pursuing in itself, but it is not going to get us one inch closer to explaining the origin of language.
The authors also remark: “It has been suggested that the presence of functionally flexible vocalisations in pre-linguistic human infants is evidence for an evolutionary divergence towards speech production.” In casual speech, remarks of this sort are okay, but evolution does not move “towards” anything. You can say tautologically that a trait appeared and, as it was the sine qua non of a later trait, the later trait became possible. Nothing more profound follows, and it certainly does not follow that bonobos have taken a turn toward speech.
The authors conclude: “In summary, the current study contributes promising insights into the evolution of human speech by suggesting an intermediate stage between fully-fledged functional flexibility in human speech production and the more traditionally viewed of fixed signals of non-human primates.” Authors often pat themselves on the back for having done more than they did.
Comments