I read the studies in a book in 1994. Here is something I found on the net with similar information:
Some conclusions and opinions of linguists are:
William Samarin has written: "When the full apparatus of linguistic science comes to bear on glossolalia, this turns out to be only a facade of language — although at times a very good one indeed. For when we comprehend what language is, we must conclude that no glossa, no matter how well constructed, is a specimen of human language, because it is neither internally organized nor systematically related to the world man perceives." 18
J.G. Melton 14 wrote briefly of Samarin's findings, who concluded that glossolalia is not a true language. Only a few consonants and vowels appear in it.
An academic Internet mailing list, "The Linguist List" focused on glossolalia in early 1995. 15
- Some of the subscribers noted that glossolalia had a simple primitive structure, and exhibited much repetition of individual sounds.
- One commented that the words spoken within a given church tended to be similar, and unlike the sounds heard within in another congregation.
- Another commented that his observations among American churchgoers showed that they "seem to latch onto and then repeat sounds that sound foreign to them, and intersperse the name 'Jesus' in between the sounds."
- Still another said that: "there are two continental charismatic traditions - a French one concentrating on melodious spontaneous song and a German/English one concentrating on speech."
- A subscriber stated that "Some years ago as an undergraduate, I memorized the first eleven lines to Beowulf. Occasionally I recited them to people (I still do). Once I recited them to a friend from Alabama, and she told me that if I did that back where she came from, folks would say I was speaking in tongues."
- The moderator noted that the: "native language of the speaker was a pretty good predictor of the kinds of sounds that would occur in glossolalia; one general pattern was that sounds perceived as generally marking "foreign" speech (whatever that may mean) would occur, while sounds perceived as typical of the native language would not. Thus, for American English speakers, /r/ would be rendered as the alveolar trill, never as the American retroflex; on the other hand, these speakers would not include the low front vowel in their glossolalia, /ae/-as-digraph, because that's perceived as a typically "American" sound for some reason. On the other hand, truly exotic sounds--those not typical of the native language, but that don't happen to be familiar to speakers of the language--would tend not to occur: American English speakers don't produce clicks in their glossolalia."
- D.J. James quotes some conclusions of William Samarin: "When the full apparatus of linguistic science comes to bear on glossolalia, this turns out to be only a facade of language — although at times a very good one indeed. For when we comprehend what language is, we must conclude that no glossa, no matter how well constructed, is a specimen of human language, because it is neither internally organized nor systematically related to the world man perceives."
One analytical study of glossolalia was performed by an unknown person or persons. 16 An individual's ecstatic speech was tape recorded and played back separately to many individuals who sincerely and devoutly believed that they had received the gift of interpreting. Their interpretations were quite inconsistent. e.g. one said that "the utterances referred to a prayer for the health of someone's children." Another interpreted the speech as "praising God for a recent and successful church, fund-raising effort." It is obvious from that study that those particular interpreters were unable to extract significant meaning out of the glossolalia. However, they were probably not conscious of that fact.
http://www.religioustolerance.org/tongues1.htm