Last week it was widely reported that a Croatian girl awoke from a brief coma speaking fluent German, to the surprise of her mother and the doctors. The case was hailed as a mystery and “medically unexplainable.”  The story spread mainly through those paper with less than a stellar journalistic reputation – but today that matters much less than it used to. Now stories get on the internet, regardless of their source, and become part of the seamless viral spread of information.

There actually isn’t a legitimate foreign language syndrome, and superficially the story is highly implausible. (Incidentally, there is a foreign accent syndrome, but this is really just a speech deficit that causes altered pronunciation that happens to superficially resemble a specific accent.) Croatian is a Slavic language, but has some German words mixed in due to its history with the Hapsburg Monarchy. The reports indicate that the girl studied German in school – and so she could speak it before her coma. Not much of a mystery, really.

We discussed the story on the SGU, prompting a listener from Croatia to write in with more details (local media always seems to have more detailed and accurate reporting). Ivan Osman wrote to me:

Now the interesting part is that after she woke up from the coma she could speak German a lot more fluently than before and not a word in Croatian. My guess is that Sepsis caused a brain damage in left temporal lobe. Probably mostly in Broca’s Area, thus disabling her in speaking Croatian but not damaging her knowledge of German language. At that point her brain probably switched to best alternative and her passive knowledge of German sprang to life. Without possibility to fall back to Croatian vocabulary there is no dilemma in which words to choose and how to use them so her German must have sounded a lot better to doctors and her parents. In my opinion the most unsettling thing about this story is not mystical nonsense around it but a quote from a neuropsychiatrist that was connected to this case. He said “This case is medically unexplainable” – only a day after the whole incident. Seems to me that people like him should not be even remotely connected to science. If he is capable of abandoning any rational explanation only a day after girl woke up. I could understand if he spent 50 years researching on that particular case and/or related cases and then states he can’t explain it. But his reckless statement made this story a mockery of journalism and his career instead of interesting medical case what it probably is.

The new detail here I had not previously read is that she stopped speaking Croatian. The story was distorted in the English press so that the interesting point seemed to be that she was speaking German, which is not interesting because she already could speak German. They missed the really interesting point – not that she could speak German, but that she could not speak Croatian.  Ivan brings up an interesting question – can you neur0logically lose your primary language, and retain (and even improve) a secondary language?

This situation is called bilingual or polyglot aphasia. Aphasia is a language deficit due to brain injury. For some quick background – the major language areas include Broca’s area (posterior dominant frontal lobe), which is responsible for translating language into speech – essentially for the subtle and agile control of all the muscles of speech. Damage to the Broca’s area produces what we call non-fluent speech – it is broken, hesitating, and stuttering. But if this were the only area of the brain damaged comprehension would be intact, and the content of speech would be normal. A Broca’s lesion would also not discriminate among languages, and so Ivan’s hypothesis can be rejected.

The other primary area is Wernicke’s area (posterior superior dominant temporal lobe) which is responsible for translating ideas into words and words into ideas – this is our dictionary. Damage here results in impaired comprehension of language. Speech is fluent, but the content is mixed up, and may contain neologisms (inventing new nonsense words), letter and word substitution, and generally nonsensical speech. There is also the arcuate fasciculus, which connects Wenicke’s to Broca’s. An isolated lesion here will impair repetition of words that are heard, but comprehension and fluency will be intact.

It should also be noted that when a second language is learned after around the age of four, then other parts of the brain are used for that language function in addition to the language cortex – but the primary language cortex is still required. It is currently not clear if languages learned together from infancy (compound bilinguals) are encoded in identical places in the brain or separate, although there is some evidence that they may not entirely overlap. This could allow for the differential damage of one language over another.

Generally, with aphasia (damage to any part of the language cortex) polyglot languages are lost and recovered together (in parallel). Also generally speaking, secondary languages (especially those learned after age four – called coordinate bilinguals) are more impaired and recover slower than primary languages.

There are various kinds of non-parallel polyglot aphasia reported. One language may be more severely lost, or recover later than another. There may also be involuntary switching between languages (inserting words from one language when trying to speak the other). And cases have been reported where a patient could only speak one language at a time, sometimes on alternating days.

To further complicate matters is the question of speech therapy. Most often speech therapy given in either language will improve speech in all languages. However, the primary language is usually chosen for therapy – the first language learned, or the dominant language currently used. There are cases, however, of differential recovery – recovery only in the language that is receiving speech therapy.

Neuroscientists are still trying to make sense of all this. One proposal is that in polyglots, while speaking one language, the other language(s) have to be inhibited. If this ability to inhibit one language is impaired, then this will result in involuntary switching or mixing up the languages. It is further proposed that some patients with aphasia may have the inability to turn off the inhibition of one language – so the language may still be there, but permanently inhibited and unavailable. Such cases are rare but provide a window into the organization of multiple languages in the brain.

Getting back to our Crotian girl, as I stated above I doubt she has a Broca’s lesion. She also likely does not have a dense Wernicke’s lesion either, as this would impair comprehension in every language. But there is other cortex involved in language, so-called association cortex involved in processing input and output from the primary language cortex. And language memory is also more widely distributed, especially for secondary or multiple languages.

The question is – could she have an unusual form of aphasia that is impairing her ability to disinhibit her Croatian language, leaving her only able to speak German? This could theoretically have the effect of making her German seem more fluent, because she does not have to expend mental energy inhibiting her Croatian – that has become automatic. This would be doubly rare (perhaps unique) because Croatian is her primary language, and German her secondary language.

This story has suddenly become more interesting. I would like to hear more details about the case, like how old she was when she learned German. It is, of course, possible that her language change is more psychological and neurological. But in any case she deserves a detailed evaluation by a speech expert and the case should likely be reported in the technical literature (not just the lay press, who managed to mangle it).

I hope her doctors do not persist in the silly notion that her case is “unexplainable” – there is actually a literature on this question and neuroscientists are making progress in sorting out how multiple languages are processed in the brain. This case, if properly explored, could provide a valuable addition to that literature.

This case may also demonstrate, therefore, why the scientific and critical approach to anomalous cases is more valuable than the mystery-mongering (“unexplainable”) approach. It is better to assume that we can figure things out, if we are willing to try.