Blog Rating

Selected Books by Edmund Blair Bolles

  • Galileo's Commandment: 2500 Years of Great Science Writing
  • The Ice Finders: How a Poet, a Professor, and a Politician Discovered the Ice Age
  • Einstein Defiant: Genius vs Genius in the Quantum Revolution

« Apes are Shrewd Listeners | Main | Step by Step from Apes to Shakespeare »


J. Goard

I'm still confused as to why so many theorists take a vocal-auditory origin of language as the most likely scenario. As a linguistics student who dabbles in semiotics and evolutionary theory and has studied ASL, I'm at least convinced that gestural origin should be considered the primary working theory, if not the sole plausible theory.

I base this mainly upon five points, roughly in descending order of importance IMHO:

(1) The gestural-visual channel allows for vastly more iconicity, not merely for entities but, much more importantly, for grammatical phenomena like transitivity, aspect, and referent tracking/anaphora. Once mostly iconic signs are established in a population, principles of historical change naturally explain the progressive development of symbolic communication.

(2) In our nearest surviving relatives, manual dexterity, voluntary manual control, manual communication in the wild, and learning ability for manual signs in the lab, are all much greater than vocalization capacity (the latter radically so).

(3) Early acquisition of fluent sign language appears to be universal given exposure, despite the absence of signing as a significant selective environment in recorded history. Genetic history of deafness does not appear to make someone a better signer than those deafened pre-linguistically by illness or injury.

(4) Evidence such as the McGuirk effect suggests that human phonetic perception involves not merely sound discrimination, but calculation of a speaker's physical gesture. Anything like that in macaques??

(5) All spoken languages employ gesture as a significant support, despite its amazing neglect by much linguistic theory.

Ed, do you think the vocal-auditory origin story is more plausible?
BLOGGER: The system reported on this week does not assume any particular modality. It works the same way for signs, sounds, and scents.

As to my own particular opinion, I find all the arguments for signs first as unpersuasive, and suspect that from the beginning sound and gesture have been used together. I need a good explanation for why and how, if sign language was already in place, we gained voluntary control over our vocalizations and replaced one modality with another.

J. Goard

I can see that the general modelling method could be applied to speech or sign, but my problem is with the comparison the authors make (and their choice of title!!!), referring to the semiotic structure of a much more distant relative's vocalizations, rather than to the chimpanzee's manual and facial gestures. Why?

I actually don't think the problem you mention is the biggest one for the sign theory. (At least not the "why" part.) Assuming that a full-blown manuo-visual language was in place, the selective pressure behind increased control over vocalizations (and development of a vocal tract which can produce more diverse vocalizations) seems obvious, viz. the frequent need to communicate in conditions of visual impairment, or while using one's hands.

To me, the big problem for the sign origin view is archeological. If hominids had full-blown language long enough to drive the physical changes involved in oral language, it seems like complex artifacts should go back much further. In response, I guess I have to suppose that my intuition about the relation between symbolic communication and lasting technology is seriously flawed.
BLOGGER: An interesting post. I wouldn't give up on the sign-first argument just because you think the archaeological evidence should be stronger. David Gil's work certainly calls into question the relation between complex technology and complex language.

But one caveat. Beware of the "obvious" evolutionary benefits, especially in language. They hide many problems in plain sight.

Paul Strand

Skinner analyzed the functions of language in his most important book, Verbal Behavior. He indicated that similar processes are at play at both ontogenetic and phylogenetic levels--selection by consequences. The idea that selection is operating at the ontogenetic level with respect to language acquisition has been vituperously rejected by linguists. But more and more evidence points to the fact that the acquisition of language requires that it be functional for the individual and not just the species (See this issue of Adaptive Behavior and also the work of Ernst Moerk).

An expanded and data-driven analysis of how selection drives language acquisition is provided by Doug Greer of Columbia University. In an article cited below (among others by him), functional stages of language acquisition are discussed. Greer shows how certain functional verbal units must be in place for others to emerge. (These arise out of experiences within an environment in real time).

Functional units like those described by Skinner and Greer are inconsistent with the units arising from a structural-linguistic analysis. Unlike a structural analysis, however, 50 years of functional analytic work has given rise effective language acquisition interventions.

Studies of the functions of language show that most individual differences among humans regarding language abilities likely stem from motivational issues (i.e., differences with respect to what is and is not reinforcing to an individual at a particular point in their development). These may reflect biological or environmental deficiencies. Regardless of their origin, teaching language to non-verbal individuals (i.e., those with deficiencies such as autism) requires an understanding of those units and also the conditions for creating acquired reinforcers. (Certain acquired reinforcers motivate verbal behavior because they make it useful. Once it is useful, language use and skill advances).

For language to emerge in a species or in an individual, it must be functional.

Greer, D. & Keohane, D.D. (2009). The evolution of verbal behavior in children. Journal of Speech-Language Pathology and Applied Behavior Analysis, 4, 10-39. (retrieved March 15, 2010 from:


Blogger says,"I need a good explanation for why and how, if sign language was already in place, we gained voluntary control over our vocalizations and replaced one modality with another.", in his answer to first comment.
I do not think that signing has been replaced. We all do it to some extent. I have known in my life two people who could not use their hands to do something (like wash dishes or draw a picture) and talk at the same time. As soon as they started to talk, they stopped their manual task and then started gesturing. If they went back to the manual task, the speech got slower and died.
If gesture and speech are linked in the fabric of the brain and also are found together in behaviour,then why would the idea of a connected origin seem unreasonable?
BLOGGER: We need an explanation for how the human lineage developed voluntary control over its vocal cords. We have direct links between the brain and our vocal apparatus. Other primates don't. Developing the control needed for speech was not a trivial issue.

I suspect that all this speech-first vs signing-first dispute is a false issue. From the beginning I believe we were gesturing, speaking, and making faces.

Paul Strand

Terrence Deacon (1997) brilliantly noted that properties of language that make its acquisition easier for children are selected for over the course of evolution. This may explain the advantage that vocal speech has over signed speech. That is, as noted by Greer and Keohane (2009, p. 24): “The capacity to hear what one reads [or says] is important because the acoustical physical properties of sound allow more “bits” to be transmitted by sound than is possible with signs. For example, children who are deaf from birth have extreme difficulty developing reading comprehension beyond Grade 6 (Karchmer & Mitchell, 2003).”

Studies of deaf children make it clear that vocalizing does more for humans than allow adults to communicate over longer distances. Vocalizations allow children to be their own listeners in a manner that is more efficient and more rewarding than what is possible through signing.

The process of becoming a listener to one’s own speaker behavior seems to be a key to human cognitive capacities that are lacking in other species. For example, speaking to one’s self is a strategy that both adults and children use to increase their ability to resist short term temptations and stay focused on larger, delayed rewards.

Research on primates illustrates that they can acquire many aspects of language through teaching, including syntax. What they appear to lack, however, is the motivation to perform these acts. These motivational deficiencies likely stem, in part, from the fact that primates are limited to sign language as opposed to vocal language. This may be a source of the difficulty of teaching primates the full panoply of language, just as it is an apparent impediment to allowing deaf children to reach their full linguistic potential. The deficit is likely motivational and stems from the informational poverty of signing compared to vocalizing.

Raymond Weitzman

To Blogger:

Can you explain your statement "We have direct links between the brain and our vocal apparatus."? Animals also have vocal apparati. Aren't the links between their brains and their vocal apparati direct? If not, what is the difference between in the linkages? How does one determine neurologically whether there is voluntary control or not?
BLOGGER: It will take me a little time to address this issue is accurate detail, but I willl do so. My silence right now does not mean I am ignoring the question.


Paul: You are seriously misinformed I'm sorry to say. There is no 'informational poverty of signing compared to vocalizing', and it is not true that the 'physical properties of sound allow more “bits” to be transmitted by sound than is possible with signs'. If anything the opposite is true. Nor can the language problems of deaf children be attributed to a purported deficiency of visual languages, outside of sociological factors.

Unfortunately I can't access your reference(s?), but welcome you to contact me, skwiver at gmail dot com if you would like to pursue the matter.


"We have direct links between the brain and our vocal apparatus." (Blogger) “Animals also have vocal apparati. Aren't the links between their brains and their vocal apparati direct?” (Raymond Weitzman). Can we differentiate human vocal productions from primates’ calls and birds’ songs? I would focus on motor imitation, or more concretely, learning of new and complex articulatory-phonetic patterns. At a certain age, the child, during her first reproduction of a learned pattern, manages to move beyond trial and error, and, virtually within days, also successfully performs a delayed ‘first reproduction’. These two simultaneous achievements can be explained - this is Piaget’s hypothesis - by means of a pre-motor plan or latent imitation that the child would have carried out while observing the model. By contrast, Marler’s experiments reveal that in birds, dialectal learning is a very different matter. Given that young sparrows store the adult singing during their mute and learning stage, Marler deprived some sparrows of hearing just when they were about to start singing, and thus proved that learner birds, in order to become singing adults must first hear their own singing attempts, detect to what extent these are similar to or dissimilar from the model and rectify accordingly. (This, incidentally, can be extended to the abilities of parrots.) In short, there is no pre-motor plan or latent imitation that the bird would have carried out while hearing the model. Probably latent imitation is a very demanding process. Raymond Weitzman has posted a good question. I do not know the answer, but I have wanted to highlight its importance.

Paul Strand

Dear Lyssa,

I was able just now to access the Greer article via the link that I posted above, but I am sorry you were unable to.

Another way to access it is via Once there, click on the link for Journal of Speech-Language Pathology and Applied Behavior Analysis. Then click the Past Issues link. Then click Volume 4. Scroll down to article.

Best of luck accessing the article.


Paul Strand

Correction: That should be

My apologies,


Paul; thank you, I was able to access the article. Its statement that the acoustic medium can transmit more data than the visual is given with no evidence, as a casual aside in a discussion limited to aspects of the reading process. The parenthetical “[or says]” was not part of the cited material. The claim is simply not true. Reading has no bearing on language evolution, and postulating an 'informational poverty of signing' merely attempts to revive the myth that some languages are primitive, as anthropologists used to say. A couple of other points:

II. Speech never replaced vision, it only augments it. Gesture, including facial expressions, are a sociolinguistic variable that co-occurs with spoken language. Either component can be voluntarily suppressed to fit conditions, within limits.

III. No evolution of the VT should be necessary. All that is required for language is a dozen or so perceptually distinct combinatorial units. The fingers of one hand, or one's face, are capable of several times that. The Hawaiian Language uses only thirteen such units, and Bonobos produce 15 such units acoustically. They also combine these in different sequences, partly freed from context.

Apes lack the same voluntary control of the VT that they have for the face and body and use to communicate via visual signals. Noble et al seem to ignore this matter of voluntary control. It is far easier (and phylogenetically prior) to establish the speech communicative triangle with vision than it is with speech. OTOH visual languages require R to voluntarily follow S's gaze/point to establish the triangle, and to voluntarily maintain it afterward. The acoustic medium only requires voluntary action on the part of S. I suspect selection pressure would favor the latter in later stages and the former earlier.

Paul Strand

Dear Lyssa,
I agree with you that with respect to an analysis of representational capacities, sign languages are in no way inferior to vocal languages. There are, however, informational advantages to vocalizing over signing that are not accounted for by such an analysis, having to do with behaving verbally in the world. These examples are obvious and not so obvious. On the obvious side, one can reach a wider audience with vocal behavior—an audience that cannot see you. In this way, vocal verbal behavior has a higher informational value to the extent that it can affect the behavior of a larger number of people.
A less obvious way has to do with how we experience our own verbal behavior. As I noted in an earlier post, this involves listening to one’s own speaker behavior (i.e., talking to one’s self). When I vocalize to someone, I experience the same sounds they do. When I sign with my hands, however, I do not have the same experience as someone across from me—I do not see my hands from the same position as they do. This is a difference in the shared experience of my speaker behavior. In addition to this, what I mentioned above with respect to communicating with others holds for self-talk too; I am limited to signing to myself when I can see my hands. Also, I cannot talk myself through a difficult problem while engaging my hands elsewhere, as so many adults and children do when they are faced with a challenging task.

How might these differences affect verbal capacities? Let me start by saying that it is true that the passage I quoted from Greer and Koehane (2009) has to do with reading behavior. But reading is an important part of verbal behavior. In addition there is much evidence in the literature dating back decades that deaf individuals have substantially lower Verbal IQs than do non-deaf individuals. Importantly, however, their Performance IQs are about equivalent to those of non-deaf individuals (see Remine et al., 2007, for example). This patterning of IQ findings suggests that what is absent for deaf individuals is a level of verbal input and interaction—at an interpersonal and personal level—that is not absent for non-deaf individuals. And this absence has serious implications for verbal behavior development.

As I noted in my previous post, this relates to the larger discussion of language in non-human primates. It suggests that disappointments with respect to teaching them verbal behavior may stem, in part, from the limitations of signing.

Remine, Maria D.; Brown, P. Margaret; Care, Esther; Rickards, Field (2007). The relationship between spoken language ability and intelligence test performance of deaf children and adolescents. Deafness and Education International, 9, 147-164.


This blog is not the place to discuss problems Deaf people have with speech. Hearing people don't have those problems, and deaf hominids in the Pleistocene were most likely happy meals.

You may not agree but people can and do sign without seeing their hands; they sign in their sleep; blind people sign; and, when reading a signed language one mentally produces the words on the page, with little or no perceptible movement, the equivalent of silent reading or subvocalization. The reading process you mention is identical for sign and speech, and the 'self-listening', the perceptual feedback essential for language, takes place at the phonological level, in the left hemisphere of the brain, where there is no distinction between sound or vision.

Sound accesses an audience that can't see you, but vision accesses an audience that can't hear you. This cuts both ways. Alarm calls are well suited to give warnings, but other communicative functions are better served by facial expression and body language, and that is how apes use them. The communicative toolkit of any species contains sound, smell, gestures, facial expressions, and more. It is undoubtedly simplistic and naïve to select one, ignore all the others, and say our language, with its complex interplay of gesture and speech, is attributable to that one factor only.
BLOGGER: "when reading a signed language one mentally produces the words on the page." That's an interesting point. Thanks.

The comments to this entry are closed.

Bookmark and Share

Your email address:

Powered by FeedBlitz

Visitor Data

Blog powered by Typepad