There’s a longstanding debate over the intellectual capabilities of Homo erectus. As we’ve discussed before, below the neck erectus is highly similar to modern humans. It’s average brain size is lower than that of modern humans, but it’s still within the range of modern human genetic variation. In the past we’ve discussed the possibility that they used boats and engraved shells. The archaeological evidence is sparse, so there’s lots of room for speculation and disagreement. Now, a new paper in the journal Biological Theory, “Did Homo Erectus Have a (Vocal) Language?”, reviews “a comprehensive set of evidence in favor of vocal linguistic capability in Homo erectus.”
Brain Size
The first thing that comes to mind in the context of intellectual capabilities is brain size. The paper notes that “brain volume for most of the Homo erectus fossils is above 800 cm3, varying from that level up to 1390 cm3, which is within AMH [anatomically modern human] variation.” They continue:
Compared with brain volume, encephalization quotient (EQ) may be a more reliable predictor of intelligence. It is calculated as brain mass in grams/(0.056×body mass in grams0.746). As pointed out previously, Homo erectus’ brain volume varied regionally and temporally, and therefore its EQ possibly varied accordingly. Take, for example, Homo erectus fossils excavated at Zhoukoudian, China, and dated to 400 to 500 kya or even 680 to 780 kya. Their brain volume ranged from 980 to 1075 cm3, and their body weight averaged about 47 kg. If we focus on 980 cm3, the lowest level of brain volume, the EQ is ~5.70; if we choose the highest level of 1075 cm3, the EQ is ~6.29, which is very close to the EQ (6.48) for humans and higher than either the EQ (5.89) for Neanderthals or the EQ (4.25) for early Pleistocene Homo erectus from Africa. Therefore, as far as intelligence is concerned, at least some groups, if not all, of Homo erectus could have had a sufficient level of intelligence to support language, especially when compared with Neanderthals, whose EQ is 5.89 and whose vocal linguistic capability has received increasing evidence.
In other words, brain size should not preclude Homo erectus from speaking. They even found that “at least some groups of Homo erectus soon after its emergence had a human-like frontal lobe.”
Vocal Anatomy
Of course, if you’re going to talk then you need the right vocal anatomy. Here again, the paper argues that Homo erectus had the right vocal anatomy for speech:
The descent of the larynx and the hyoid is among the anatomical traits appealing to research on flexibility and complexity in vocal communication. It is a salient trait in humans, compared with nonhuman primates. It arguably resulted from bipedalism and was therefore observed in Homo erectus. The descent led to an approximation of the ratio of the horizontal and vertical proportions of the supralaryngeal tract to 1:1, which is typical of the human type.
They also cite evidence that erectus lacked laryngeal air sacs, another human-specific trait that may be necessary for speaking. Moreover, they found that erectus had “a spinal canal within the human range of variation” and propose: “Homo erectus might have had a similar level of neural control for speech to that of modern humans if the spinal cord is an effective predictor.”
Regarding hearing, they write that “most features found in later humans were already present in Indonesian Homo erectus” at 1.4 Ma, suggesting “the hearing apparatus of some groups of Homo erectus could have been able to cope with some form of language.”
Genetics and Population Genetics
The problem with genetic evidence is that no one knows for sure exactly what genes or mutations would produce human-like cognitive and speaking abilities. I don’t put a huge amount of stock in these genetics arguments, but the authors put much emphasis on the gene FOXP2 as being important for human language. Here, they argue that Homo erectus probably had the same FOXP2 gene that we modern humans do:
FOXP2 has been regarded as the most relevant gene to language-related vocal control. FOXP2 in modern humans differs from the non-human primate version in that human FOXP2 contains two amino acid substitutions, a property also shared by Neanderthals, suggesting that these substitutions occurred before the split between Homo sapiens and Neanderthals. According to Diller and Cann’s (2009) calculation, the two amino acid substitutions occurred 1.8 to 1.9 mya shortly after Homo erectus emerged, rather than in the past 200 thousand years, as postulated by Enard et al. (2002). The substitutions in FOXP2 could have contributed to vocal control and learning, as well as to the evolution of human speech, by adapting cortico-basal ganglia circuits. In addition to its contribution to spoken language, FOXP2 was responsible for anatomical adaptations related to bipedal locomotion. Therefore, bipedalism, an emergent trait typically seen in Homo erectus, reversely indicates that FOXP2 had been well placed by the time of Homo erectus.
They also make a population genetics argument of sorts based upon introgression, the transfer of genes from one population to another. The argument goes like this: Homo sapiens can speak, and Homo sapiens interbred with Neanderthals and Denisovans, so those other groups could probably speak as well. They argue, “Therefore, it makes sense based on vocal linguistic capability in Homo sapiens to assume that both Neanderthals and Denisovans, and even the common ancestors of all three species, could speak.” They then want to see if Homo erectus can be added to this happy speaking family, asking: “[H]as there been any evidence available for introgression between Homo erectus and any of these later species of Homo?” They propose that the answer is yes:
In sum, evidence is indeed available for introgression between Homo erectus and later humans if we consider the flexibility and diversity of Homo erectus as an evolving species, so that the so-called archaic or superarchaic species are included as its varieties. Thus, it is reasonable to assume vocal linguistic capability in at least some groups of Homo erectus by using the previous inference model.
It is indeed an inference-based model, and if it holds then it supports speech in erectus.
Archaeological Evidence
Lastly, they argue that evidence for “symbolic thinking … entailed the emergence of linguistic symbols,” which suggests speech. They cite evidence for symbolic thinking in Homo erectus in its ability to make Acheulean stone tools, their seafaring ability, the use of controlled fire, and engravings made on bones. Regarding Acheulean stone tools, they argue:
[E]xperiments on the making of Acheulean bifacial axes showed that, compared with the control group that was instructed by demonstration only, the experimental group that learned through demonstration combined with verbal instruction performed much better because speech enabled them to better conceptualize the particular sub-goals that pertained to the sequences of the actions the teacher performed. If the requirement for hierarchical planning and symbolic communication in Acheulean stone toolmaking is coupled with Homo erectus’ human-like brain region of BA44, which supports syntactic processes as well as action, and with the approximation of its vocalization and hearing abilities to those of AMHs, the argument for a vocal language in Homo erectus becomes more convincing.
They further argue that producing such complex tools would have required linguistic communication:
Acheulean toolmaking involved more complexity in structural planning and execution. The cultural transmission of this kind of toolmaking skills presupposed processes of teaching and learning, which would have been impossible without cooperation between the teacher and the learner. As argued by Gärdenfors and Högberg (2017), such cooperation could hardly have been achieved without linguistic communication.
As for boat-building, they again argue that producing such complex tools would require cooperation, symbolic thinking, and the ability to speak:
Archaeological findings on the island of Flores in Indonesia indicate that Homo erectus voyaged to the island and settled there as early as 850 kya, from which it follows that Homo erectus was capable of making raft-like voyage tools that necessarily required symbolic thinking and linguistic communication.
They also note that the use of controlled fire implies symbolic thinking, and “Evidence for the controlled use of fire by Homo erectus comes from the fossils excavated from Zhoukoudian, China, and from the site of Wonderwerk Cave, Northern Cape province, South Africa.”
Summary of the Argument
The conclusion to their paper offers a summary of the evidence they believe supports speech in Homo erectus:
Homo erectus is regarded as a crucial turning point in the evolution of features typical of humans. Homo erectus was the first Homo species to have a greatly expanded brain. Its brain features showed great similarity to those of AMHs, including EQ as an indicator of intelligence, the way of brain development, the morphology of brain regions (for example, frontal lobe and parietal lobe), and some important cerebral connections. The anatomical features of vocalization and hearing displayed no obstacles to processing some form of language. Genetically, many important mutations related to brain, cognition, and language could be traced back to the time of Homo erectus, and introgression between Homo erectus (or its varieties) and later humans has also been vindicated. Symbolic thinking attributed to Homo erectus is noticeable from its tool production and use. Cooperative toolmaking, hunting, and infant breeding have testified to an advanced level of cooperation among Homo erectus, which, as argued by some researchers, requires symbolic communication.
They admit that the evidence is circumstantial, but they believe it is “collectively convincing” that Homo erectus used linguistic communication.