One of the most notable theoretical physicists of the twentieth century, Erwin Schrödinger, considered it “obvious” that there is only one human consciousness, and that the feeling of having an individual mind is just that, a feeling (Schrödinger 1944). With all due respect to the father of the wave equation of quantum mechanics (for which he received the Nobel Prize in 1933), I will continue to assume that we all have our own minds and our own self-awareness, but only one of these.
It is not difficult to imagine the Viennese genius turning in his grave if he were to read the paper “Two languages two minds” (Athanasopoulos et al. 2015), because if the authors are right, an extra proliferation of minds in people is involved. Athanasopoulos et al. echo Charlemagne’s well-known claim that speaking another language is like possessing another soul, and they update the claim by implying that speaking two languages is like having two minds. The phrase works well as a headline (although not as suggestive as Charlemagne’s), but implies a rather drastic devaluation of what we usually mean by mind.
The conclusion that speaking two languages is like having two minds is based on the interesting fact that speakers of different languages tend to group events differently. Athanasopoulos et al. point out that monolingual German speakers are more likely to identify a video in which a person walks towards a car (without seeing whether the person reaches the car) with a video in which someone enters a house, than with a video in which someone goes towards a distant building without reaching it. This result would imply that, when in doubt, these speakers tend to focus more on the end of the event (entering the house) than on its development (walking towards the distant building). The authors conclude that German speakers categorize the (incomplete) car video as of the first type. The reverse situation occurs with monolingual English speakers, who tend to identify the car video with the distant building video (focusing on the unfinished development of the event). According to the authors, this is so because the progressive aspect is grammatically encoded in English, while it is not in German. According to their data, only 37% of English speakers identified the car video with the house video, compared to 62% of German speakers. This is undoubtedly an interesting phenomenon, one worthy of an explanation, but in my opinion it does not justify the assertion that a “German mind” is different from an “English mind”.
Bilingual subjects are the tricky case here. If a German mind is different from an English mind (let us assume so for now), what is the mind of German-English bilingual speakers like? Or do they have two minds?
In a first experiment described in the article, German-English bilinguals tended to behave like monolinguals, though with less pronounced differences. So, when the experiment was conducted in a German context (in that the subjects are given instructions in German and they have to speak in German), there is a greater likelihood that subjects will choose the house video as being more similar to the car video (and not to the distant building video), whereas when the same subjects are placed in an English context, they tend to choose the building video (which marks the event in progress, not completed) as being more similar to the car video. Although the authors do not say so, this more moderate behavior of bilinguals can be an effect of the interference between languages, so that when they speak German they are less “German” than German monolinguals, and when they speak English they are less “English” than English monolinguals.
The second experiment is more interesting, but less statistically conclusive, and more cumbersome in design. In this experiment the “verbally mediated categorization” of bilingual speakers is hindered by making them repeat aloud a sequence of numbers while performing the same tasks from the previous experiment. According to the authors, this interference task inhibits the language in which it is assumed that the subjects are thinking, and leads them to choose the options that correspond to the other language. So, when bilinguals are doing the experiment in a “German context” and their German language is interfered as they are asked to repeat numbers (in German) while watching the videos, they are less likely to choose the house video (the one preferred by German speakers), and are more likely to choose the far building video (the one preferred by English speakers). Halfway through the experiment, a change is made, so that they are now asked to repeat numbers in English (supposedly to disrupt access to this language) and at this point they tend to make choices more like “Germans”. The explanation offered by the authors is that when the task of verbal distraction interrupts access to the language of the context of the experiment, the other language takes over, so to speak.
This is not the place to assess the assumption that access to language (or, as the authors say, to “verbally mediated categorization”) can be interfered by the task of verbally repeating numbers. But it is worth noting the apparent contradiction in assuming that the active language can be interrupted, and assuming at the same time that the other language, supposedly inactive, comes to supply the “disrupted” one, because then it is not clear that the verbally mediated categorization (whatever it may be) has really been interfered. Note that the same task of distraction applied to monolinguals would imply that no language is used for the test (because they have no other language that may be supplied here), which simply shows that the influence of a given language in categorization is modest, unless we assume that in such a case there is no categorization of any kind. But this makes no sense, since in that case the monolingual subjects would not be inactive, but would continue identifying the videos (although using something that is not language, according to the logic of the authors).
The paper claims to have shown that English-German bilingual speakers have a flexible categorization of the world depending on which language is dominant in a particular task of identifying motion events. Its conclusions are rather ambitious: “These findings show that language effects on cognition are context-bound and transient, revealing unprecedented levels of malleability in human cognition.”
But if the effects of the spoken languages on cognition are context dependent and transitory, I think that what this study shows is that the effects on cognition of the specific language one is using are really weak or superficial, so that speaking of an unprecedented malleability in human cognition is not appropriate. If we accept such a claim, then we would also have to acknowledge that the term cognition is used in a different sense when used in the quoted text and when used to refer to human cognition in general (as compared, for example, to feline or avian cognition).
Anyway, the suggestion that bilinguals have two minds contrasts with much research on bilingualism in recent years. Kroll et al. (2015) summarize this progress in a comprehensive state of the art review, which concludes: “Contrary to the view that bilingualism complicates the language system, this new research demonstrates that all of the languages that are known and used become part of the same language system”.
Bilinguals (contrary to what was believed in the past) do not work as two monolinguals, but their languages interfere with each other and tend to coalesce into a single system of knowledge. It has even been observed (see Kroll et al. for references) that the brain tissue employed in the storage and processing of the two languages is essentially the same. Of course, the bilingual brain is different from the monolingual brain, but not because the bilingual brain is split into two systems of knowledge (let alone two minds), but because it develops a more complex system of knowledge whose management increases certain abilities, just like lifting weights every day makes your biceps grow.
The most notorious finding in recent decades of bilingualism research is that the two (or more) languages are always active and interfere with each other. It does not matter if the L1 is very dominant over the L2, if the two are very different in their morphology, phonology, syntax or even orthography, if one is signed and the other oral, or even if only one of them is used regularly. The knowledge of a second language continuously and incessantly affects the use of the first language, and, of course, the knowledge of the first language (much more robustly) affects the use of the second one. The possible cognitive benefits of bilingualism derive from the extra need for the bilingual speaker to inhibit one of the languages when using the other, which provides an apparent improvement in the ability of solving cognitive conflicts (both in the use of language as elsewhere) and even increases the protection against certain types of cognitive degeneration, including a delay of symptoms of Alzheimer’s Disease. Actually, the fact that the two languages are always active and interfere with each other is to be expected, in that they are part of a single system of knowledge (language). This conclusion does not support the vision of “two languages, two minds”, but, on the contrary, it shows that if there is a part of language that structures our mind (and it would be very strange if this were not the case), that part coincides with what languages have in common.
For the reader to be convinced of how frivolous (in the use of the word mind) the assertion that a bilingual has two minds might be, it is worth considering an example of the opposite case: a person with one language and two different minds. Indeed, the question of whether there are people who have two minds (that is, if they are more than one person) has been seriously discussed, and I do not refer to cinematographic cases of dissociative identity disorders, but cases of people with a section of the corpus callosum connecting the two hemispheres of the brain. Perhaps the most famous case is P.S., a boy with a split brain studied by Gazzaniga and collaborators (LeDoux, Wilson and Gazzaniga 1977).
Although it is a controversial issue, the authors suggest that each of the hemispheres of P.S.’s brain were self-conscious and had their own mind. The interesting thing (in terms of what interests us here) is that after surgery, only the left hemisphere could talk, but both understood speech, and the right hemisphere began to communicate by assembling Scrabble letters to form words, using the left hand. It seems that P.S. had, unlike patients examined thus far, a substantial part of language in the right hemisphere (even though he was not left-handed). The exciting fact is that the right hemisphere was shown to have feelings, to know what day tomorrow was, what profession he would like to have (and this was different from the one he had said when using the left hemisphere) and, in general, all the attributes of a human mind. As LeDoux, Wilson and Gazzaniga noted, “each hemisphere in P.S. has a sense of the self, and each possesses its own system for subjectively evaluating current events, planning for future events, setting response priorities, and generating personal responses.”
The most important conclusion is that the fact that only in the case of P.S. is the cognitive independence of the right hemisphere detected, while the right hemispheres of other patients do not reveal such capacity for self-awareness (with the possible exception of another patient named Vicki), suggests to the authors that “the presence of a rich linguistic system is a reliable correlate, and perhaps a necessary prerequisite, to some of the richer aspects of mental life.”
It thus seems that the ancient intuition that language lies behind consciousness and the very nature of the human mind makes sense. But, contrary to the suggestion of Athanasopoulos et al., the ingredients that form the fabric of our human mind are not the superficial, external aspects of language that differentiate German from English (i.e. those aspects subject to historical change and, therefore, variation), but what is common to all languages, including, of course, the various languages in the (single) mind of a bilingual. I am convinced that this idea would have seduced Schrödinger.
[Author’s English version of a blogpost published in Spanish (10 April, 2015) in zaragozalinguistica.wordpress.com, and previously published in The Linguist at NTNU 3, 2016]