Linguistic Analysis Of Some Reversals

By: Pierre

Pierre has a Masters degree in Applied Linguistics, and has studied phonetics and phonology. He is also a certified Reverse Speech Analyst.

We are pleased to present here a linguistic analysis of 10 reversals on the Reverse Speech website. The first 8 reversals are linguistically sound and the final 2 reversals have some issues. More reversals will be added as time permits.

We here at ReverseSpeech.com believe that accuracy of documentation is the most important goal that we should be striving for. To this end we have developed the Reverse Speech check points that all analysts must follow (See list of check points at end of article.). If a reversal does not meet these check points then it should not be documented. Of course the Reverse Speech check points is only a guide and sound linguistic principals must be followed to check all reversals. Reverse Speech will never be taken seriously by mainstream unless sound principals of documentation are followed.

However, it should be noted that in the attempts by the unconscious to communicate, there may well be glitches (as is the case with the last 2 reversals). There are parameters and boundaries to how much distortion we should accept, depending on quality and quantity, otherwise RS may as well be thrown in the monkey cage for the monkeys. This is why it is important that people doing RS have adequate training about RS, are disciplined and scientific in their approach, are professional, firmly control biases and prejudices, and have a good knowledge of the properties of speech and sounds and their interrelationships

Reversal analysis is presented below. Click on the reversal to hear the sound file.

David Oates

ITS AN HONOR

Documented reversal	It’s an honor
Example phonemic transcription	ˈɪts ən ɑnəʳ
Phonetic transcription	ˈɪts ən ɑ̃ʾnəʳ

ANALYSIS

This is an accurate documentation of the sound segment. The articulation of the vowels and nasalization of the first vowel in honor is an excellent representation of American pronunciation.

MY ADVICE IS RANCID

Documented reversal	My advice is rancid
Example phonemic transcription	mae ædvɐɘs ɘz rænsɘd
Phonetic transcription	maɛ ædvɐɘ̘s ɘ̘z ɹænsɘ̘d

ANALYSIS

This is a very good representation of the words documented. We can see good linguistic sense at work here. The diphthong vowel in my does not rise to a high vowel as a result of the particular articulation occurring in the forward speech. But this also makes sense linguistically when we look at it from the Reverse Speech side. The influence of the low front vowel following, /æ/, in advice would have the effect of stopping the rise of the diphthong to a high vowel. We can also look at this from a different perspective – If the diphthong vowel had risen to a high front vowel, the following vowel (the initial of advice) would have probably been a centralized vowel closer to a schwa. Therefore the initial in advice in the reversal is a more natural vowel in light of the final position of the preceding diphthong.

YOU’RE FRIGHTENED, LEAN ON ME

Documented reversal	You’re frightened, lean on me
Example phonemic transcription	jər fraetnd liːn ɑn mi
Phonetic transcription	jɵʳ fɹaet̚ n̟ˠː liɪ̞n ɑ̃mniː

ANALYSIS

This is a good representation of the documented words. Even though there is no clear evidence of an actual  /d/ articulation at the end of frightened, we can see here how it is simulated through the length of the /n/ which has a simultaneous velar articulation, and the increasing pressure in the palatal area. There is evidence of some alveolar articulation at the end of the vowel and before the onset of /n/. However, a lack of it would not have changed the structure of the word and would remain fully recognizable as frightened. The lack of an /n/ in on is a natural linguistic occurrence whereby the /n/ assimilates in part to the following /m/. Also, the fact that the vowel /ɑ/is nasalized produces a natural perception of the /n/. However, there is an /n/ between the /m/ and the final vowel which is coming off the forward speech /n/. It is produced in a way in Reverse Speech that allows our perception to easily ride over it. The first two points discussed here are explainable as linguistic phenomena and the last point is a perceptual issue.

I’M SO FULL OF SHIT

Documented reversal	I’m so full of shit
Example phonemic transcription	ɐen soʊ fʊlɘ ʃɪt
Phonetic transcription	ɛ̆e̞n so̞ʊ fʊlɘ ʃɪt

ANALYSIS

This is a good representation of the documented sentence, even though /m/ doesn’t occur in I’m. The existence of an obvious /n/, however, doesn’t detract from a clear understanding of the sentence. Perhaps it was unnecessary for the unconscious to interfere and influence at this point because of this. As well, in speech, the fact that /s/ is an alveolar articulation would mean that its place of articulation would likely mean that some influence over a preceding consonant occurs to assimilate to the same position (which /n/ has), particularly in moderate to faster speech. From the onset of I in I’m the tongue is traveling upward through the vowel glide to the sibilant /s/. We can deduce here that it was probably unnecessary for the unconscious to interfere in order to influence production of an /m/, as not producing /m/ in this case doesn’t really detract from clear understanding.

I’M NOT TELLING

Documented reversal	I’m not telling
Example phonemic transcription	ɐɪm nɒt tɛlɪŋ
Phonetic transcription	ɐ̞ːnɑɒt̚ tˢɛlŋ

ANALYSIS

The sounds are appropriate representations of the words I’m not tellin’. The child produces the reversal in a British accent. There is a couple of things to note – the vowel sound in not starts with the unrounded vowel immediately followed by the rounded version of the same vowel (The vowel sounds are a little higher than the standard IPA sounds, but as the diacritic that I want to use is unavailable on Lucinda Unicode I have not put it in). There isn’t an /m/ before the /n/, however, and the lips come closer together after the /n/ in line with the partially articulated /m/ in the forward speech. However, this doesn’t affect clear recognition. In the forward speech English but I’m not we can note the assimilation of the /m/ in I’m to the /n/ in not. In fact, if the /m/ had been pronounced more strongly by the speaker it may have affected negatively the good quality of the reversal. There is another reversal option here. The missing /m/ could indicate that it is Are not tellin’.

HEAL MY FRIEND

Documented reversal	Heal my friend
Phonetic transcription	hɪɘɫ mae fɹe̞n

ANALYSIS

The documented reversal is a good representation. As can be heard behind the forward speech there is an extra syllable produced at the end which may give the impression of there where the th assimilates to /n/ before it. It presents some issue, however in light of the good quality of heal my friend, it should be considered safe to document this.

I’D DO IT AGAIN

Documented reversal	I’d do it again
Phonetic transcription	aɘ̘dʉ̙wɪ̙t ɪgɛn

ANALYSIS

The selected audio section is representative of the words I do it again. It is questionable whether there is alveolar contact by the tongue preceding the production of the /d/ in do. However, this is not such a serious point. There is one real issue here in the phonemics. In listening to the longer section of forward speech in reverse, there is sk at the beginning and this is not separate from the I. However, it may be that this part is separate to the rest and it is actually do it again. In the forward speech Edwards is speaking about relentless negative attacks against John Kerry and that this will continue in the weeks ahead. So Edwards is not talking about himself and what he would do personally. There are two ways to interpret do it again. First, it just confirms the expectation that Kerry will continue to be attacked. Second, it represents a hidden mischievous aspect of Edwards that doesn’t always have the best wishes for Kerry. One must be very careful about considering that the second interpretation is true and watch that one doesn’t just desire it to be true. This is a danger of interpretation and analysis. There is yet another issue that should be considered, that it represents a constant, that is, it is typical behind the specific forward speech words. That is a real danger one must be aware of. If it is typical, then the unconscious may not be manipulating or influencing the speech. Therefore whatever interpretation you lend it is irrelevant.

THEY CANNOT KILL IN ANY FORCE. SEE THAT PRESSURE

(Rumsfeld, at the beginning of the Afghanistan campaign)

Documented reversal	They cannot kill in any force. See that pressure
Phonetic transcription	d̻ɪ̞kɛnɐxkɪlɘnə̘nɪfɔəs si̞d̻ɛʔt̚pɹe̞ɪ̞ʂɹ

ANALYSIS

The reversal on the whole is fairly sound. They alters shape in normal running speech and is therefore reasonable here. The vowels in the other words are all within an acceptable range. We see the assimilation of the /t/ in cannot to the velar following it, which is a normal occurrence in forward speech. There is a strong impression of /f/ in force, however it may be that the sound is only being caused by the non-verbal noise of the recording. This could, therefore, be a case of phonemic restoration, a perceptual phenomenon. The constriction in the throat at the onset of the vowel may also help to give the impression of a consonant. However, the sheer number of phonemically acceptable segments, the good prosodic quality and correct grammar of nine words makes this a strong speech reversal. I found this reversal at the start of the Afghanistan campaign in late 2001. The U.S. government was trying to make the Northern Alliance effective against the Taliban and I remember that there were media articles that week about the frustration of the US leadership towards it. I would say that the reversal refers to that pressure.

INDIAN REPORTER: THAT MAN IS JUST AN IDIOT

Documented reversal	That man is just an idiot
Example phonemic transcription	ðæt mæn ɪz ʤʌst ən ɪdɪət
Phonetic transcription	ʥæd mɛlɘb ʤ̊ə̙tˢθ əne̞ɪdɪɛ̙n

ANALYSIS

Considering normal running speech, that and just are acceptable. There is frication occurring in the initial of that and the length of the vowel indicates a /d/ ending rather than /t/. The initial in just has a degree of devoicing and there is an unclear frication occurring at the end. However, in light of the speaker’s Indian accent and the facts of running speech, they are acceptable here as representations. Man is however, possesses a clear /l/ without indication of an /n/ influence. There is no is. An idiot is problematic. There is a lack of evidence of a /t/ final while the /n/ final is quite clear (from behind the first /n/ in indian). When an idiot is uttered in forward speech there is generally a constriction occurring in the throat/back of oral cavity after an and before idiot. However, that doesn’t happen here. There is also a diphthong vowel occurring after the /n/ in an whereas the initial /i/ in idiot would have a single vowel only. The fact that there are three issues present puts it in doubt. The closest possible word would seem to be Canadian. However, there is a lack of supporting evidence of a velar initial /k/, yet at a certain point it may be perceived. I have not put it in the transcription.

THEY PULLED OUT THE GRASS

Documented reversal	They pulled out the grass
Example phonemic transcription	ðeɪ pʊld ɐʊt ð grɑs
Phonetic transcription	(ð)ðe̞ɪ pʌɘ̘ɪ̙ˈlɜɘ̘ɪ̙ pʰf̊ brˠɐɑzs(p̚)

ANALYSIS

Listening to the reversal gives us the perception of the five words. However there is a greater complexity of sounds occurring that is not perceived through general listening. We can get the perception of a /d/ in pulled out, because of its strength and precise position after the previous sounds, however on closer listening it bears the quality of a strongly articulated /l/. The vowel sounds in pull out are not adequately similar to the target vowel sounds. However, their glide movements can give the impression of the words on less focused listening. The strong acoustic intensity of the vowel sound in out ‘drowns out’ the following sound and can give the assumption that it is the. However the sounds are closer to a /p/ but with the lips being drawn in a little and tensed with a slight aspiration followed by an /f/ like sound. An impression of a velar /g/ sound in the last word can occur as a result of the strongly articulated /r/ that has some velar articulation itself. However, if you forget for a moment the impression of a /g/, there is also a /b/ before the /r/ (from before in the forward speech). Therefore it could just as well be bras. One might think that if it ends in a voiceless sibilant as the word grass does, then that would make it sound like brass. However, there is voicing of the sibilant occurring before it becomes voiceless. The question is, is the unconscious acting on this vocalisation, and if so how is it doing it? Has the unconscious influenced the speaker to produce a higher acoustic intensity at certain sections to create the perception of language?

Reversal Check Points

The following list of reversal check points have been compiled to give students easy steps to distinguish between gibberish, imagination and a genuine speech reversal. If followed meticulously, their use will greatly improve the student’s accuracy rate. To use them follow the following procedure.

First of all find a suspected reversal and then go over the phrase several times applying the following checking factors. Document the reversal along with the check points that it meets.