The Wikipedia page made a passing mention that people of different backgrounds "speak in tongues" differently, but I was curious if there were any papers on this topic. I'd be interested in if there is some correlation between how people try to speak gibberish, and their actual native tongue. Can anyone point me to a paper on the subject, or give some details on this?

3 Answers 3


Here’s a paper by Heather Kavan from Massey University: “We don’t know what we’re saying, but it’s profound”: The language and contexts of glossolalia. I found this paper to be enlightening, and if I have time some of her references seem worth it to follow.

Some of her points:

  • Glossolalia is no different than asking someone to invent a language on the spot and speak it.
  • Utterances contain real words that are rendered in disguised form. In other words, the speakers are applying all sorts of phonological changes to actual words to produce nonsense.
  • Open syllables (CV) are the most common. (My guess is that they are the easiest to produce.)
  • In churches, the people are given “encouragement” to speak in tongues.
    • Things that are uttered may ultimately be borrowed from other people.
    • People have rehearsed.
  • A minority of Christian groups had what the author terms “spontaneous glossolalia” (as opposed to context-dependent glossolalia):

    Spontaneous glossolalia is an intense uprush of vocalisations where tongues flow without prompting or prior knowledge of them, and the participant does not know what is happening to them. This was experienced in the purification group, it was evident in a minority of Christian accounts, and is likely to have occurred during the early Pentecostal revival.

There was another document I read a while back that was intriguing, but I do not know which of my browsers’ history to look to retrieve the URL.


There are a few published papers, but not many, and a prominent problem with such studies is that they are not carried out by experienced linguists, except in the case of William Samarin. The above-mentioned paper by Kavan is apparently not published and is not at its original link. William Samarin published a book entitled Tongues of men and angels (1972: MacMillan) and an article in Hartford quarterly (8: 49-75) entitled "The linguisticality of glossolalia" – I don't have either of these. Michael Motley published an article "A linguistic analysis of glossolalia: Evidence of unique psycholinguistic processing" in Communication quarterly 30:18-27. Osser, Ostwald, MacWhinney & Casey published "Glossolalic speech from a psycholinguistic perspective" in Journal of Psycholinguistic Research (2 (1): 9-19). then there is this MA thesis Internal-external locus of control in glossolalics by J Coulson.

The lack of published acoustic corpora is a prominent problem in these studies. The reason why this is significant is that such phonologically-relevant generalization as are presented by the researchers are based on their judgments of phonetic content. Accurate phonetic transcription is an arcane art not widely practiced in linguistics, and to the extent that professionals have any prior experience in phonetic transcription, it is usually heavily biased in favor of English phonemes. Without accompanying acoustic corpora, it is impossible to know what phonetic details were omitted, and we cannot track the general reliability of the researchers as field workers (except Samarin), because they are not linguistic field workers.

In this regard, it is of interest that Motley reports the presence of non-English sounds in his corpus, and gives sufficient descriptive detail that I am inclined to take his data claims at face value. He noted the presence of non-English segments [β ɲ x] and apico-dental [t d n] in one of the glossolalic varieties (of the bi-glossolalic informant) as opposed to apico-alveolar in his English. It is also observed that there were many non-English clusters such as [fw, vr, wx, dl]. At least for this speaker, I would conclude that the glossolalic output is not just a filtered version of "possible English".


The “tongues” Pentecostal and Charismatic Christians are producing today is an entirely self-created phenomenon. It is non-cognitive non-language utterance; random free vocalization based upon a subset of the existing underlying phonemes of the speaker’s native language, and any other language(s) the speaker may be familiar with or have had contact with.

It is, in part, typically characterized by repetitive syllables, plays on sound patterns, alliteration, assonance, and over-simplification of syllable structure. It is also interesting to note that any disallowed sound combinations, i.e. consonant clusters, in the speaker’s native language are also disallowed in his/her tongues-speech. Further, this subset of phonemes typically contains only those sounds which are easiest to produce physiologically.

Sometimes the same speaker will draw on different subsets; thus claiming to speak in "diverse tongues".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.