Lecture | Louis-Jean Boe & Thomas Sawallis | Which Way to the Dawn of Speech?
Which Way to the Dawn of Speech?
Reanalyzing half a century of debates and data in light of speech science
Louis-Jean Boë1 & Thomas R. Sawallis2
1 GIPSA-lab, CNRS, Grenoble Alpes University, France
2 New College, University of Alabama, Tuscaloosa, USA
In the weeks around New Years, 2017, two complementary articles discussing speech evolution appeared in respected general science journals: Fitch et al., 2016, in Science Advances and Boë et al., 2017, in PLOSOne. These two articles announced the final failure of a theory that had been widely propagated and broadly accepted for half a century, despite numerous critiques and partial falsifications: the laryngeal descent theory (LDT) of Lieberman and colleagues (Lieberman, 1968; Lieberman et al., 1969; Lieberman & Crelin, 1971). Taken together, those studies represented – and continue to represent – an extremely powerful research paradigm, drawn directly from the core understanding of speech science, that the acoustic speech signal transmits articulatory information. Specifically the authors: (i) used recorded calls of live primates to make anatomical inferences about vocal tracts (VTs), (ii) used either VT casts of extant primates or reconstructions of fossil species’ VTs to make acoustic inferences, and (iii) by appropriate comparisons of anatomy and acoustics, drew conclusions regarding both the ontogeny and the phylogeny of speech.
Their conclusions, later termed LDT, were: that fully human speech, in particular the full human vowel inventory, was made possible by the large pharyngeal cavity resulting from laryngeal descent, which occurs over the lifespan of anatomically modern Homo sapiens (AMHS) only; that living primates, pre-modern humans (including Neanderthals), and modern human toddlers were restricted to schwa-like vowels; and that speech could only have developed after the emergence of AMHS some 200,000 years ago, and language more recently still.
Controversy in speech evolution research is inevitable, due to lack of fossil evidence, difficulty of experimental design and data collection, absence of general paradigms, and especially, the need for multidisciplinary cooperation among otherwise compartmentalized fields. We will give some taste of that controversy as we trace 3 decades of difficult work showing that a large pharynx from laryngeal descent is not necessary to produce the full inventory of vowels. Theoretical arguments claimed that infants, Neanderthals, and primates were anatomically able to produce contrasting vowels, and recorded evidence accumulated that infants could as well, and finally, the articles noted above presented MRIs of macaque VTs and especially recorded calls from baboons showing both produce contrasting vowels, all without the large pharynx required by LDT. If we think of this evidence as in some sense “fossils” of early speech emergence, this pushes the “dawn of speech” 100 times further back in our history, to our last common ancestor with old world monkeys, over 20 million years ago.
We will present new evidence we have recently developed further reinforcing that claim, and will outline certain implications for language evolution theory & research more generally.