When did humans start talking? There are nearly as many answers to this perplexing question as there are researchers studying it. A new brain imaging study claims to support the hypothesis that language emerged long before Homo sapiens and coevolved with the invention of the first finely made stone tools nearly 2 million years ago. However, some experts think it’s premature to draw sweeping conclusions.
Unlike ancient bones and stone tools, language does not fossilize. Researchers have to guess about its origins based on proxy indicators. Does painting cave walls indicate the capacity for language? How about the ability to make a fancy tool? Yet, in recent years, scientists have made some progress. A series of brain imaging studies by Dietrich Stout, an archaeologist at Emory University in Atlanta, and Thierry Chaminade, a cognitive neuroscientist at Aix-Marseille University in France, have shown that toolmaking and language use similar parts of the brain , including regions involved in manual manipulations and speech production. Moreover, the overlap is greater the more sophisticated the toolmaking techniques are. Thus, there was little overlap when modern-day flint knappers were making stone tools using the oldest known techniques, dated to 2.5 million years ago and called the Oldowan technology. But when knappers used a more sophisticated approach, called Acheulean technology and dating to as much as 1.75 million years ago, the parallels between toolmaking and language were more evident. Stout and Chaminade have used functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) scans, although not on the same subjects at the same time.
In the new work, published online today in PLOS ONE, archaeologist Natalie Uomini and experimental psychologist Georg Meyer, both at the University of Liverpool in the United Kingdom, attempted to advance these earlier studies in several ways. They applied a technique called functional transcranial Doppler ultrasonography (fTCD), which measures blood flow to the brain’s cerebral cortex and which—unlike fMRI and PET—is highly portable and can be used on subjects in the field through a device attached to their heads (see video above). The fTCD approach makes it much easier to monitor subjects’ brains during vigorous activity, such as the somewhat violent motions that are required to make stone tools. Uomini and Meyer are also the first to study both toolmaking and language tasks in the same subjects.
The researchers recruited 10 expert flint knappers and gave them two different tasks. In the first, the knappers crafted an Acheulean hand ax (see photo), a symmetrical tool that requires considerable planning and skill. The procedure involves shaping a flint core with another stone called a hammerstone. While wearing the fTCD monitor, the knappers worked on the tool for periods of about 30 seconds each, interspersed with control periods of about 20 seconds in which they simply struck the core with the hammerstone without trying to make a tool.
In the second task, the knappers were asked to silently think up words beginning with a given letter. The control periods consisted of simply resting quietly and not thinking of words.
The team found that the pattern of blood flow changes in the brain during the critical first 10 seconds of each experimental period—when the knappers were strategizing about how to shape the core or thinking up their first words—was very similar, again involving areas of the brain implicated in manual manipulations and language. Moreover, although there were some variations in the patterns between the 10 knappers, the toolmaking and language patterns within each individual were very closely aligned —suggesting, the team concludes, that the same brain areas recruited in both tasks.
The results, Uomini and Meyer argue, support earlier hypotheses that language and toolmaking coevolved, perhaps beginning as early as 1.75 million years ago. This doesn’t necessarily mean that early humans were talking in the same rapid-fire way that we do today, Uomini points out, but that “the circuits for both activities were there early on.”
Stout calls the new study “exciting work” that provides “one more piece of evidence supporting a link between stone-tool making and language evolution.” Yet a number of questions remain, he says, such as whether the correlation is between the motor skills involved in making tools and in making the sounds of speech, or whether toolmaking and language share higher cognitive functions such as those used in symbolic behavior .
That question is critical, some researchers say, because the knappers in this study and the ones that Stout conducted probably used a technique known as the Late Acheulean, dating from about 500,000 years ago, which put a much greater emphasis on symmetry and aesthetic considerations than did the earliest Acheulean, dating from 1.75 million years ago. “There is an enormous difference” between these varieties of Acheulean toolmaking, says Michael Petraglia, an archaeologist at the University of Oxford in the United Kingdom, who adds that “future experimental studies should thus examine the range of techniques and methods used.”
Thus the new work is “consistent with the hypothesis” of coevolution between language and toolmaking, “but not proof of it,” says Michael Corballis, a psychologist at the University of Auckland in New Zealand. “It is possible that language itself emerged much later, but was built on circuits established during the Acheulean” period.
Thomas Wynn, an archaeologist at the University of Colorado, Colorado Springs, is even more cautious about the results. He thinks that the fTCD technique, which measures blood flow to large areas of the cerebral cortex but does not have as high a resolution as fMRI or PET, “is a crude measure, even for brain imaging techniques.” As a result, Wynn says, he is “far from convinced” that the study has anything new to say about language evolution.