Old habits die hard
Claims about language universals have been abundant throughout the history of linguistics. The subject of this claim has ranged from something as broad as the structure of language (Chomsky, 1980) to something as specific as a single universal word (Dingemanse et al. 2013). However, this claim has rarely been expressed more strongly than in a 2016 newspaper headline which read: “Humans may speak a universal language, say scientists” (Knapton, 2016). Such a bold statement merits critical investigation, and below I will compare the statements in this article to both the research that spawned it, and the wider context of linguistic research.
The newspaper article in question was published on the website of the UK paper The Daily Telegraph and was written by the paper’s science editor Sarah Knapton. This final point is relevant because this author has previously been criticised for misleading headlines regarding the risks of vaping (Bates, 2016; Rodu, 2015). This criticism apparently went as far as two formal complaints to the UK’s Independent Press Standards Organisation (IPSO). IPSO subsequently ruled that Knapton did break the Editor’s Code of Practice on one occasion (Rodu, 2015).
Returning to the subject at hand, I will now briefly summarise the findings of the academic paper on which Knapton’s article was based. Damián Blasi and colleagues (2016) found that, for a limited list of words, statistically robust associations exist between the concepts of those words and the sounds that are used to verbalise them. By controlling for genealogical and geographical relations between languages, the authors ensured that these associations reflected similarities between unrelated languages. The authors do not make any hard claims about the cause of these relationships, but they suggest that “sound symbolism, iconicity, communicative pressures, or synaesthesia” might play a role (Blasi et al., 2016, p. 10821).
Nowhere in the original research paper do the authors claim that “humans may speak a universal language”. The attribution of this claim to the scientists, then, is clearly a factual inaccuracy by Knapton. Furthermore, the claim itself is not supported by any research whatsoever. In addition to this flagrant exaggeration, the main text contains some smaller inaccuracies that will now be discussed.
Perhaps the second-most important misrepresentation of the facts in Knapton’s article concerns her suggestion that the fundamental nature of the concepts that were investigated is somehow related to the finding of universal sound-meaning associations. In her words: “as if concepts that are important to the human experience somehow trigger universal verbalisations” (Knapton, 2016). However, it should be made very clear that the researchers only looked at basic concepts. This means that we have no way to tell whether more specific words do not have similar concept-sound combinations across different languages.
Knapton also makes some mistakes in reporting some of the more specific findings. For example, she states that “Other words found to contain similar sounds across thousands of languages include ‘bite’, ‘dog’, ‘fish’, ‘skin’, ‘star’ and ‘water’” (Knapton, 2016). However, a closer look at the results show that these similar sounds are shared across hundreds rather than thousands of languages (Blasi et al., 2016).
Finally, the author also misrepresents the researchers in a more direct way. She claims, for instance, that some of the researchers are from the Netherlands or the US, even though none of them are Dutch or American. It could be that Knapton is referring to some of the authors’ academic affiliations with Dutch and American universities. However, this would be inconsistent as she refers to the Argentinian nationality of the first author in the same sentence. Furthermore, at one point she refers to Morten H. Christiansen as “Dr Christian”.
To Knapton’s credit, she does include a brief statement by a linguist who calls the reported research into doubt. However, this expert’s main criticism of the research, namely that universal sound-meaning relations could be explained by “a common ancestor language” (Knapton, 2016) has actually been accounted for in the design of the study. This does not mean that the reported research is infallible, however. For example, neither the journalistic article nor the academic paper point out that the decision to compare abstracted symbols of sounds rather than the physical acoustics of sounds potentially biased their analysis to find universal sound-meaning associations. It is widely acknowledged that certain abstracted sound units (such as [r]) have wildly different acoustic characteristics across languages (Ladefoged & Maddieson, 1996, pp. 215-216).
In sum, the biggest problem with Knapton’s article is the headline, as it clearly misrepresents the findings of the scientists whose research is reported on.
- Bates, C. (2016, August 12). Telegraph science editor Sarah Knapton puts the record straight. Not really. Message posted to The Counterfactual blog: https://www.clivebates.com/telegraph-science-editor-sarah-knapton-puts-the-record-straight-not-really
- Blasi, D. E., Wichmann, S., Hammarström, H., Stadler, P. F., & Christiansen, M. H. (2016). Sound–meaning association biases evidenced across thousands of Languages [Supplemental material]. Proceedings of the National Academy of Sciences, 113(39), 10818–10823. http://www.pnas.org/cgi/doi/10.1073/pnas.1605782113
- Chomsky, N. (1980) On cognitive structures and their development: A reply to Piaget. In M. Piattelli-Palmarini (Ed.), Language and learning: The debate between Jean Piaget and Noam Chomsky (pp. 35–52). Cambridge, MA: Harvard University Press.
- Dingemanse, M., Torreira, F., & Enfield, N. J. (2013). Is “huh?” a universal word? Conversational infrastructure and the convergent evolution of linguistic items. PloS one, 8(11), e78273.
- Knapton, S. (2016, September 12). Humans may speak a universal language, say scientists. The Daily Telegraph. Retrieved from http://www.telegraph.co.uk
- Ladefoged, P., & Maddieson, I. (1996). The sounds of the world’s languages. Oxford: Blackwell.
- Rodu, B. (2015, July 23). British Media Watchdog Faults The Telegraph for Demonizing E-Cigarettes. Message posted to Tobacco Truth blog: http://rodutobaccotruth.blogspot.nl/2015/07/british-media-watchdog-faults-telegraph.html