Talk:Phonetic palindrome

"Stops in a sunny spot" is a good example of an English phonetic palindrome.


 * I'm afraid it isn't in my accent, where I would pronounce it /staps ɪn ɜ sɜni spat/, which backwards would be /taps in ɜ sɜnɪ spat/, or "Tops een a sunnih spot." Besides the s, the problem is with the difference between the "i"s in in and sunny. A great try, though, I'd bet it's very close to one… $$\sim$$ Lenoxus " * " 22:40, 21 July 2008 (UTC)
 * This is wrong. The phrase would have to be "Tops in a sunny spot" to work. RoachPeter (talk) 07:57, 12 July 2017 (UTC)

I found a fun one in the 80s, and now I have a place to share it. "Yeah right" sounds the same when reversed. Kilorat (talk) 10:05, 15 August 2011 (UTC)
 * Not in my accent, it doesn't. RoachPeter (talk) 07:57, 12 July 2017 (UTC)

Scientific evidence needed
This article contains quite a few examples, and it's regrettable that most are not supported by sound files. There is one overwhelming flaw in the whole article, however, and that is the unquestioning use of the concept of utterances “sounding the same”. Anyone who has experimented with synthetic speech knows that after you have played your bit of sound material to yourself many times, it can be clearly heard as saying what you intended it to sound like, but if you ask someone else to say what they hear, the answer is usually not what you intended. This is a form of auditory illusion. Psychologists have worked on such illusions for a long time. The “Verbal Transformation Effect” is perhaps the best-known: hearing the same thing over and over again often results in successive perceptions of a whole series of words, some phonetically quite dissimilar (I have been a subject in a VTE experiment, and found the effect quite disorienting). There is a Wikipedia article on Semantic satiation that describes a somewhat similar illusion. My feeling is that if you equip yourself with some software that can play a sound recording backwards, and you then start listening to any old scrap of speech material played forwards and backwards, pretty soon you will convince yourself that you are hearing something different from what you heard at the outset. And if you are looking for bits of speech that sound the same whether played forward or backwards, you will soon convince yourself that you have found some. This article is essentially anecdotal, and badly needs some scientific evidence to justify it.

-- RoachPeter (talk) 10:04, 5 July 2017 (UTC)
 * In the article Palindrome there is a section headed 'In speech'. I now want to propose that the article 'Phonetic palindrome' be merged into that section and removed from the list of Wikipedia articles. As part of that process I want to remove a lot of errors about phonetics and phonemes, and to make it clear that what is claimed about perception of phonetic palindromes has no basis in scientific testing. Please reply if you object to the proposal. RoachPeter (talk) 15:04, 7 July 2017 (UTC)
 * This gets more complicated. In addition to the Phonetic palindrome article that I have been criticizing for having no scientific evidence to support its statements, there are also articles on Phonetic reversal and on Reverse speech. All three describe claims about speech perception that are based purely on anecdote and speculation, with no proper scientific evidence. At least the Phonetic reversal and the Reverse speech articles acknowledge that the claims are unscientific. There must surely be good reason to merge all this material into a single article. In addition, it all needs to be thoroughly revised to prevent readers from thinking they are being presented with proven facts. I should also mention the article on Backmasking, which describes the use of reversed recorded speech in music, but this is written to a much higher standard, describing the use of Backmasking from a historical and musicological point of view but not making claims about its psychological or phonological aspects. If a multiple merge is (as I suspect) too complex a thing to undertake, I return to my basic point, that the article on Phonetic palindrome is not good enough to stand alone as a WP article and should be merged into Palindrome. RoachPeter (talk) 13:43, 8 July 2017 (UTC)
 * I have started to make improvements and corrections to this piece, prior to recommending merger into Palindrome. I have cut the lead shorter and made a new section on contextual effects so that the difference between forward and backward can be illustrated with a sound file and spectrogram. RoachPeter (talk) 15:48, 11 July 2017 (UTC)
 * My biggest concerns with phonetic palindromes is that they are dependent on the pronunciation of the person speaking, and we have virtually no sources for this phenomenon. What is a phonetic palindrome for one person may not be for someone else, and we have no verifiable reliable sources. Unless we can find solid sources for this I would suggest getting rid of this article. It has been tagged for 6 years, and the online sources appear to be copies of Wikipedia. Meters (talk) 18:31, 11 July 2017 (UTC)

I've undone the merge. This was not an uncontested merge, and there were major improvements made byuser:RoachPeter to the content just hours before the merge. We need to look at the new content and see if a merge is warranted.

I still see no solid references that discuss phonetic palindromes. Where does this term come from? Most of the hits I'm finding appear to be simply rephrasing of the content of this article or are even Wikipedia mirrors. If this term is something that a Wikipedia editor made up then this is WP:OR. The article was completely unsourced for several years after it was created by User:Lenoxus so maybe he or she can provide some insight.

Even the simple examples we use in the article are described as having differences and (hence are not truly palindromic). Most of the phonetic palindrome section appears to be negative, in that it is discussing unsourced or unsubstantiated claims. If the best we can come up with is "untested claims that a piece of recorded speech 'sounds the same”'whether played forwards or backwards have to be treated with a degree of scepticism"' then we have a problem. Where are the tested claims? Meters (talk) 01:50, 5 October 2017 (UTC)
 * I would question that what you have done is undone a merge. It seems to me that you have deleted the content of 'Phonetic palindrome'. You may be right that this stuff is so unscientific that it is beyond redemption, but removing it without discussion is not helpful after all the work I did to improve it. RoachPeter (talk) 16:46, 5 October 2017 (UTC)
 * Perhaps you should look at the history before making accusations. I simply  undid the merge. User:Jytdog  removed the content and then chose to remerge what was left. Meters (talk) 17:35, 5 October 2017 (UTC)
 * My apologies - I did not look carefully enough at the history of what has been done by whom. RoachPeter (talk) 11:56, 6 October 2017 (UTC)
 * There was a ton of unsourced content as well as WP:OR/WP:SYN that I removed, and then I merged what was left per the merge suggestion. Wikipedia is not the place to publish original research.  If you want to build this out again I suggest starting at the other article, adding only content that is supported by reliable sources, and if it gets too big i am sure no one will mind splitting it out again.  Jytdog (talk) 18:06, 5 October 2017 (UTC)
 * All I was trying to do was rework the 'Phonetic palindrome' article so that it was not so embarrassingly bad, and then try to get the improved version shifted into 'Palindrome'. The whole point here is that "sounding the same" is not something that can be just established by anecdotal quotations. For a scientific study, objective testing is needed, and that is conspicuously absent from the material as it stood. Let's assume that the article 'Phonetic palindrome' is now dead and buried. What we currently have under "In speech " in 'Palindrome' is a first paragraph that is, I think, acceptable as far as it goes but does not talk about the importance of allophonic variation. The second para reproduces from the earlier article the stuff about bits of William Burroughs, which is scientifically worthless - if you listen often enough to recorded speech (especially if it is badly recorded) you can persuade yourself that you are hearing almost anything. The Michaelson examples may or may not work in a given accent, but are of no use if they are just armchair examples that haven't been tested on listeners. I have already spent far too much time on this stuff, so I will just leave it as it is. RoachPeter (talk) 11:56, 6 October 2017 (UTC)

For what it's worth, RoachPeter is a renowned professor of phonetics, so it would be less of an issue if he just wrote what he added in the article on his own blog, or of course more preferably in a peer-reviewed medium, and we used that as a source per WP:RSSELF. There still could be the problem of WP:NPOV, but that would at least render the article free of original research. Nardog (talk) 06:59, 8 October 2017 (UTC)