I agree that the continuity issue is nothing new -- that was my reply to the second note by Alex. But the most important issue is a recent article by Dehaene and his colleagues about their research with cognitive modalities other than language. That was the topic of my first note to Alex. In my note below, I'm forwarding my first note plus a diagram of the arcuate fasciculus, which was missing in my second note.
Dehaene's article shows the importance of multiple methods of thinking and reasoning other than linguistic. This makes the term ''language of thought" misleading." That is important for interpreting Peirce's writings about reasoning in diagrams and even "stereoscopic moving images."
Although Dehaene did not mention the arcuate fasciculus, I believe that it is important for explaining the difference between their results with baboons vs humans. The human fasciculus has many more links to a wide range of areas in the brain for those different ways of thinking. The chimpanzee fasciculus has fewer connections, but still quite a few more connections than the monkeys. (See the attached arcuate.png)
That may explain why chimpanzees can learn a fair but incomplete version of human sign language, but they can't go farther. Dehaene and his colleagues also analyzed infant abilities. There might also be some relevance to your studies of the Piraha language and the kinds of experiences in their environment.
Alex,
Thanks for those references. I have read other articles by Dehaene, and he does good work.
But the hypothesis of a singularity may be too strong. A more likely assumption is a continuum with baboons having a weak ability, the great apes having a stronger ability, and humans having the strongest. The article I copied below discusses different areas of the brain that are specialized for different ways of thinking. That is a well-known fact.
Other primates, including the baboons they studied, also have specialized areas of their brains. But the differences may not be a sharp singularity. It's likely to be a continuum. They should continue their study with at least one intermediate point, such as chimps or bonobos. See the attached diagram, which shows the connections between Wernicke's and Broca's areas (the two major areas for language understanding and speech generation in humans). It also shows the corresponding areas in macaques and chimpanzees.
Note that the arcuate fasciculus, which connects those two areas in humans, has many more connections to other areas in humans, fewer in chimpanzees, and very few in macaques. That suggests a continuum rather than a singularity.
John
----------------------------------
Volume 26, Issue 9, September 2022, Pages 751-766
Trends in Cognitive Sciences
Opinion
Symbols and mental programs: a hypothesis about human singularity
Stanislas Dehaene, Fosca Al Roumi, Yair Lakretz, Samuel Planton, Mathias Sablé-Meyer
Abstract: Natural language is often seen as the single factor that explains the cognitive singularity of the human species. Instead, we propose that humans possess multiple internal languages of thought, akin to computer languages, which encode and compress structures in various domains (mathematics, music, shape…). These languages rely on cortical circuits distinct from classical language areas. Each is characterized by: (i) the discretization of a domain using a small set of symbols, and (ii) their recursive composition into mental programs that encode nested repetitions with variations. In various tasks of elementary shape or sequence perception, minimum description length in the proposed languages captures human behavior and brain activity, whereas non-human primate data are captured by simpler nonsymbolic models. Our research argues in favor of discrete symbolic models of human thought.
Section snippets
A universal human predilection for symbolic structures
The Lascaux cave, south of France, is famous for its spectacular depictions of aurochs, horses, and deer, from over 18 000 years ago. A lesser-known fact, however, is that prehistoric humans also left many nonfigurative signs (see Glossary) such as rectangles, series of dots, etc. (Figure 1).
Abstract geometrical patterns are omnipresent in human productions throughout the globe and predate figurative art by hundreds of thousands of years. For instance, early Homo sapiens left a network of...
A geometric language for spatial sequences
All neuropsychologists are familiar with the Corsi block tapping task, which evaluates spatial working memory. On each trial, the experimenter taps a certain number of blocks spread in front of the participant, who has to reproduce the sequence in the correct order. In this task, working memory capacity typically does not exceed five or six locations, but this is true only for unstructured sequences. Whenever a spatial regularity is present, working memory is facilitated [22,23].
A proto-musical language for binary auditory sequences
We next wondered if the notion of nested repetition could be generalized to the domain of auditory sequences. We restricted ourselves to binary sequences using only two sounds (call them A and B). In this case, the language becomes very simple: it merely specifies whether to stay with the same item, as in the minimal sequence AA, or to switch to the other, as in AB (where the items A and B could be sounds, locations, etc.). Nested repetitions of those primitives generate long yet compressible....
A proto-mathematical language for geometrical shapes
The two previous domains involved sequences (visuospatial or auditory) and thus resembled spoken language in this respect. We next wondered if a symbolic language of thought would also be needed to account for static geometric shapes. Two tests of this idea were developed.
First, we created a static intruder test, inspired by [48], in which participants had to detect a deviant shape among five repetitions of the same base shape [49]. For instance, the base shape could be a rectangle (with...
Non-human primates fail to grasp those languages
The present work is inspired by prior proposals of computer-like languages for shapes [52,53], visual and sequential patterns [54., 55., 56.], and rhythmic musical sequences [57,58]. Its originality, however, lies in bringing this hypothesis down to simple tests that can be passed by children or adults without formal education and, most interestingly, by non-human primates. In this manner, we can evaluate the hypothesis that compositional languages are a prerogative of humans [13., 14., 15.].
Concluding remarks
We may now summarize the main aspects of our proposal. (i) Symbols, mental programs, and languages. We propose that humans are characterized by a specific ability to attach discrete symbols to mental representations and to combine those symbols into nested recursive structures called mental programs, the compositional rules of which define a language of thought. Humans develop multiple such languages of thought in various domains (linguistic, musical, mathematical…).
(ii) Conceptual productivity through...
Acknowledgments This work was supported by INSERM (Institut National de la Santé et de la Recherche Médicale), CEA (Commissariat à l'Energie Atomique et aux Energies Alternatives), Collège de France, the Bettencourt-Schueller foundation, and a European Research Council ERC grant ‘NeuroSyntax’ to S.D. We gratefully acknowledge extensive discussions with Marie Amalric, Ghislaine Dehaene-Lambertz, Joël Fagot, Naama Friedmann, Christophe Pallier, Michael Posner, and Luigi Rizzi.
-----------------------------
References from the note by Alex:
