Nov 3, 2011

Beware of speech experts bearing science

Dean Frenkel has been spruiking a new book in which he claims to have all the answers about the evolution of language. Does he? Aidan Wilson investigates.

Well over a year ago, I commented on the increasing trend of people critiquing politicians on the basis of their accent. One such person whom I missed at the time, was Dean Frenkel, a throat singer by trade who applied his singing skills to voice coaching around six years ago. He now sees himself as qualified to analyse the speech style of anyone, including politicians. As Michelle Grattan wrote for The Age in June last year:

Opposition education spokesman Christopher Pyne is a nasal speaker who seems pompous but is surprisingly effective. ”He could have an after-politics job as a voice-over for irritating commercials.”

It now appears that Dean Frenkel has been hitting the media again lately, apparently to spruik a new (self-published) book in which he expands his scholarly repertoire even further, by seeking to explain the evolutionary origins of language. Let me stress that Dean Frenkel is a singer. His closest academic involvement with the topic is that he has a graduate degree of some description in anthropology.

As part of the publicity for his book Evolution of speech, Frenkel has appeared on Late Night Live with Philip Adams, and has had another article written largely about him (ostensibly about the speech of Ted Baillieu and Daniel Andrews) in The Age. With some colleagues, I purchased a copy of said book and will duly dissect it another time. For the moment I want to concentrate on the complete nonsense that Frenkel has been spouting on radio.

The interview with Philip Adams was a chummy affair; the two appear to be close friends on-air as Adams has played his music to his ‘gladdies and poddies’ on many occasions. The entire interview can be listened to on the Late Night Live website or downloaded. The quotes below are from my own transcription and have been edited slightly to make them readable (ums and ahs have been removed, as well as false starts, mostly).

Adams begins by asking Frenkel about the theory that language has its roots in Africa.

Oh it did indeed, and very few people know this, and it’s funny, I’ve asked an audience of 300 people, where was speech invented? Amazing silence, no one answered. And when you think about it, speech was obviously invented in Africa, more than 70,000 years ago because speech travelled around the world with the first great migrants and speech wouldn’t ever have, I mean we’ve never heard of a tribe that doesn’t speak, so every single culture has speech and it has many many implications.

That language began in Africa is a fairly uncontroversial fact that is entailed by the premise that language arose at least 100,000 years ago (according to some estimates) and that Homo Sapiens Sapiens migrated from Africa more recently. I’ll ignore the fact that Frenkel describes this as ‘invention’ and concentrate on the points he makes.

Frenkel goes on to explain what evolutionary developments allowed for the ‘invention’ of speech.

About a million years ago humans became bipedal, and by becoming bipedal, i.e. standing on two legs, our voices, our voice boxes dropped and that gave us room to be able to articulate.

Adams exclaims ‘we stood up and our voice boxes dropped?’

That’s exactly right: Gravity dropped our articulator muscles so when we were on all fours, if you’ve ever tried talking on all fours, it’s actually not so easy.

There are three things that are seriously wrong with this. The first is that humans didn’t exist a million years ago; they appeared around 200,000 years ago. The second is that Homo Erectus, likely the first homonid species that lived the majority of its time on the ground in a bipedal manner, arose much earlier than a million years ago; closer to two million. The last, most laughable and possibly most egregious error is the idea that gravity was the motivating factor in the lowering of the larynx. No one really knows what caused the larynx to drop and there are several theories going around, but there wouldn’t be a scientist in the world who would attribute it to the gravitational pull of the Earth on the relatively massless larynx.

Another theory of Frenkel’s regards the evolutionary interrelatedness of singing and language. When asked what came first, singing or ‘talking’, Frenkel – a singer, remember – has no reservation in making a decision.

Without a doubt, singing. We needed to train our articulator muscles to be able to actually work and even though we had them, and singing was one of those activities that gave us control over pitch, it gave us control of the breathing involved in speech. But singing didn’t give us articulation because there was nothing to articulate, so other things like whistling like click language, which is still practised in parts of Africa, and even harmonic singing. In order to actually get every letter phoneme and word out, we needed to be able to have precise and incremental control.

I’m not going to dissect this too closely, mostly because there are a lot of garbled thoughts. Suffice to say that it shows a complete disregard for the concept of scientific rationalism and empiricism. Frenkel clearly believes singing came first because he’s a singer, but given his evidence – that we needed singing to train our articulator muscles – there’s nothing to say that it didn’t happen in exactly the opposite way. That is, perhaps we gained control over our articulators while developing language to the point where we could use them for other activities, like singing. Again, the truth is that we simply don’t know, and for Frenkel to claim that there is no doubt about the issue is either disingenuous or an indication that he is entirely untrained in this particular topic and in the scientific method in general.

Frenkel also illuminates us as to the linguistic abilities of Homo Neanderthalensis, and finally answers a question that has plagued evolutionary anthropologists ever since the Neanderthal was discovered; whether they could speak.

Well they had very much the same infrastructure as we had; they had the fundamental ability to speak. They probably had heavier tongues, and a little bit cheekily I reckon they didn’t talk, because, and they didn’t sing, because singing wasn’t macho enough. And maybe they go [grunting Neanderthal-like] Hoo Hoo Hoo and in order to turn Hoo into articulation you need very very fine movements, and there’s not other animal that I understand apart from the bird that actually has the ability of the tongue to articulate.

Yet again, Frenkel is making wild speculation about a topic that is barely explained by science, injecting into his answer complete fabrications such as the implication that Neanderthals were concerned with appearing less macho and the assumption, discussed above, that singing is a necessary precursor to language, and a gross misunderstanding of what speech is – he says Neanderthals had the ability to speak but simultaneously reckons they didn’t talk.

This last point brings up an issue that pervades everything I’ve read of Frenkel, including his book, and all his radio and other media interviews; he makes a distinction between ‘speech’ and ‘language’ but never clearly defines either. I doubt that he’s talking about ‘language’ versus ‘speech’ in the manner that Stephen Fry famously parodied in a sketch with Hugh Laurie, wherein language is to chess as speech is to a game of chess, alluding to the Saussurian division. It appears as though he intends ‘speech’ to mean any noise which uses the mouth, while ‘language’ is a matter he doesn’t concern himself with. I will have more to say about this on another occasion when I discuss his book, in which he says that “language owes its very existence to speech” (p. 6).

When Frenkel and Adams talk about more phonetically related areas, it becomes clear that he similarly lacks training, or in some cases, common sense. For example, Frenkel believes that whistling is a necessary precursor to being able to articulate the labio-velar approximant “w”.

There’s a whole stack of things going on with the lips including all the W‑words, and so um, and, if we didn’t use, and in fact Arnie [Frenkel’s solitary voice coachee], for a long time there was told that he wasn’t allowed to whistle when he was young and that meant that he wasn’t able to say any words with the letter W in it. And I was starting to try to teach him how to whistle and I got into trouble because whistling is frivolous. And so, that was one really amazing thing, and there are so many components to speech.

As an aside, around the corner from my office is the phonetics laboratory. Not being a phonetician, I thought I’d check with them before writing anything about this, and they had conniptions of disbelief. It is plainly untrue that whistling is a necessary skill to be able to produce the w sound. Think about it: how many people do you know who cannot whistle? I bet you they still have no issue saying ‘whistle’. Learning how to whistle might probably help someone learn how to articulate the w sound, but – and I’m no speech pathologist – I didn’t think that the inability to articulate the w sound was even a thing!

For the last few minutes of the interview, Philip Adams asks Dean Frenkel about the respective speech styles of several politicians, including Tony Abbott, Julia Gillard, Bill Shorten, Joe Hockey, Penny Wong and Peter Garrett. I won’t reproduce them all – listen to the interview yourself if you’re that interested in hearing Frenkel’s unqualified opinions – so I’ll stick with one.

Julia [Gillard] interestingly, she’s got two or three really unfortunate speech traits that have been exacerbated while she’s been Prime Minister. Mostly she’s actually a remarkably good speaker; she hardly ever stumbles, but those two or three problems dominate the impression, and so I constantly hear people saying ‘she’s a terrible speaker’ but it’s actually not the case. She’s mostly good, it’s only a little bit bad which makes her seem terrible.

I don’t have too much to complain about with respect to this particular point apart from that it’s an ineffectual musing from someone who doesn’t have any verifiable evidence to justify his claims.

At the heart of my dissatisfaction with this entire fiasco is that Frenkel has managed to propel himself into the media’s spotlight without so much as a shred of credibility in any of these topics. This isn’t to say he’s not an excellent voice coach; he may well be, but he’s certainly not a linguist or speech pathologist and so he should refrain from claiming to have answers to such questions as the evolution of language, the development of the vocal tract, the linguistic ability of Neanderthals or the entire field of phonetics. Phoneticians undergo years of training in order to be able to comment on phonetics, as do evolutionary anthropologists and historical linguists. Giving Frenkel and people like him airtime cheapens our entire field.

The ABC and The Age should in future ask any such self-purported ‘experts’ for their credentials, or evidence to support their claims before they put them on the air, in print or online. I’ve let Philip Adams off lightly here – I could have taken issue with his failure to adequately question Frenkel or the fact that he wrote a preface to his book. I even ignored the question Adams asked that seems to have been directly lifted from Frenkel’s press release; these things probably happen all the time in journalism and it would be unfair to single Adams out for it.

I will end by saying that I take no issue with Frenkel having opinions about politicians’ speech styles, or working as a voice coach, even publishing a book on voice coaching using the possibly novel approach of applying singing techniques. However, this does not grant him license to make huge statements that are just untrue and not backed up by any evidence.

(Visited 61 times, 1 visits today)


16 thoughts on “Beware of speech experts bearing science

  1. Lucy Cornell

    Hello Aiden,

    Thank you for your response. I appreciate your apology.

    Is there a way to remove me as the ‘stylistic segue’? I see you have also posted a link to this conversation on Philip Adams comments page on the Late Night Live website. For me, this is unnecessarily dragging on and I prefer not to have my work and name dragged through the mud with it.



  2. Aidan Wilson

    Hi Lucy,

    Firstly I will apologise for dragging you into this; my opener in this post was simply a stylistic segue. This is poor judgment on my part because I never had any issue with your credentials. My original post (in which i discuss your interview) was primarily concerned with the growing trend of critiquing politicians’ voices which, despite being a linguist, I regard as something entirely irrelevant to political discourse. That said, it’s your career and I’m not denigrating it – I fully support voice coaching. Instead I’m critical of the media that run it in place of real political discourse.

    I will also point out that I never said you were an ‘idiot’ nor do I believe it.

    I’m not critical of Frenkel’s voice coaching. I’m critical of his forays into very heavy science such as evolutionary anthropology. He has put himself out there as a self-described ‘expert’ on these matters and because of that, he leaves himself open to criticism and questions over his credentials. There’s nothing ‘glass tower’ about that.

    Let me quote myself:

    I will end by saying that I take no issue with Frenkel having opinions about politicians’ speech styles, or working as a voice coach, even publishing a book on voice coaching using the possibly novel approach of applying singing techniques. However, this does not grant him license to make huge statements that are just untrue and not backed up by any evidence.

    I stand by that.

  3. Lucy Cornell

    Dear Aiden,

    I would like to respond to what seems to have become an annual shredding of my credibility. It has just been brought to my attention now.

    Thank you for your astute critique of my words on News Radio from a year ago.

    I admit that I shamefully sound like a complete idiot in that interview.

    Please grant me the grace of it being ‘live’ and that you have transcribed what was meant for the ear into words for the eye, which have the benefit of hindsight, space and time. Still, I can see that it requires some serious attention.

    For your (and your reader’s) benefit, I would like the opportunity to discuss your premise that a degree affords you the right to make vocal assessments in public and also to assert my expertise and right to do so.

    I have a Masters Degree in Voice Research (USYD) and an undergrad degree in Arts Education: Literature/Linguistics and Theatre (UNSW). But, most predominately, I spent an additional 5 years studying in a Master/Apprentice training with Master Voice teacher, Kristin Linklater (Columbia University, USA) of which I am accredited along with only 150 others in the world. This is a practical, rigorous training in Australia and the US, the alumni of whom are esteemed, intelligent and in the top of their field in voice coaching globally. Add to that 9 years of practical experience out there in running my own business and, all up, that is 20 years in the field.

    Certainly, that interview does nothing to justify that, but perhaps the many 1000’s of people I have worked with over the years including CEO’s, politicians and leaders, internationally and at home, might.

    May I also make the distinction that I am not a speech expert and I never say I am. I respect that field too much as I do my own. My expertise lies in the voice. I am acutely aware of voice practitioners out there who are very under par and it also infuriates me that they would undermine the work that educated voice coaches do.

    So to your definition of education: I would like to assert that practical experience out in the world counts for as much, and I would argue more when combined with an education, than a degree alone. I know many speech pathologists and speech academics who are not sensitive at all to the requirements of the human behind the voice or their speech.

    I only know of Dean Frenkel. I am certain that his many years of practical experience he has establishes his right to some credibility.

    It is purely academic arrogance that might assume a degree gives you a right to an educated view.

    Now to Julia: If I had my time again to discuss her VOICE, I would say that:

    Very little breath goes into her body because (I suspect) she holds chronic tension in her throat, back of her tongue and probably in her diaphragm, pelvis and shoulders/armpits (have you seen the way she holds her body so tightly/waddles?). This tension means that the passageway for her breath is inhibited. So when she has an impulse to speak, that impulse is sent via electro-chemical impulses through her spinal chord to the muscles to do with breathing; intercostals and diaphragm. Now with these breathing muscles compromised due to tension, (which by the way most civilized adult voices are inhibited this way, hence why hardly anyone breathes fully any more) breath will not be fully drawn into her body.

    Now, in a naturally functioning voice (babies are perfect examples), a fully invested breath will release into and out from the depths of the body (literally the lungs, but there are muscles from the diaphragm that are attached into the pelvis and pelvic floor so you have a sensation of breath moving down there) up through the vocal chords creating vibrations via sub glottal pressure and in response to the nerve’s messages in the vocal chords. These vibrations are then amplified through the body ringing in and through the hard, bony surfaces of the body; chest, head, hard palate, nose etc and this is what is called resonance. The truth is that the more resonance you have the more you are able to have another human resonate with you. On a cellular level, you ring/vibrate/move another human just though the sound of your voice. This is why some people gush at Sean Connery, James Earl Jones, George Clooney, John Gielgud. Specifically for their voices, they are relying on and almost only living in their deeper notes (chest resonance). As resonating beings, humans ring/vibrate in sympathy with each other just like crystal glasses do when you tune them with your finger. So, what Sean Connery et al does (quite self consciously and self indulgently, I think), is that he relies on his deeper notes to give his audience a feeling of vibration deeper in their body… spine tingling, deeply moving, sexy and we have an experience beyond the intellect. I suspect, that was why John Laws was so successful! (I try not to fall into their trap but I know many people who do).

    So poor Julia has not got much breath going into her body. Without much breath, you start to rely on other muscles to push/squeeze/strain your voice out. When you combine that with muscle tensions around the throat, you end up with the grainy, tight, restricted, uneasy voice that she has, which is pretty devoid of resonance and so we can’t experience her energetically. What we have left to rely on is her intellect.

    Now for the argument, Cate Blanchett has a voice that rings through all the resonating chambers of her body, so we have an experience of Cate as available, transparent, disarmed and with gravitas. Why? She happily expresses herself without her habitual tensions and in turn gives us permission (on a subconscious level) to disarm and believe that she is speaking with authenticity. (She has done a lot of voice work by the way). So why mix Cate Blanchett and Julia Gillard? Because wouldn’t it be lovely to finally have an Australian politician who spoke with authenticity and intelligence, who moved us in ways that motivated us to raise our voices against the horrific issues that Australia is now facing.

    Perhaps we should follow politicians on their policies alone. But we are humans. On an animal level, we experience our fellow humans energetically and the energy of the voice is as essential to our ability to trust someone as what they are saying. It is the combination of the 2 that is the holy grail.

    Technical enough? Try saying that on radio in 10 seconds impromptu for an audience who really isn’t interested in words like sub glottal pressure, diaphragm and impulse.

    So, I do hope that this will satisfy you that I am not an idiot and that my business Voice Coach and the field of voice coaching is a valid expertise. I also hope it will help you step outside of your glass tower on your hill of academic certainty.

    And I promise to speak in full sentences in interviews from now on.


  4. A reply from Dean Frenkel | Fully (sic)

    […] following is an un-edited response from Dean Frenkel to Aidan Wilson's Beware of speech experts bearing science. I pledge in blood that Fully (sic) will be moving on to other topics soon! […]

  5. Angra

    Also AR – I agree, individual sounds which carry some immediate meaning (eg. ‘Look out, a tiger’s coming!’) do not qualify as a language. That’s the whole point of Cassirer, Searle etc.

    But I do think animal sounds have a far more complex and structured behavioural function than we may realise.

    Dolphins and whales are interesting though – they seem to even have individual names (like call signs.)

  6. Angra

    AR – interesting report on some research by Benjamin D Charlton at the University of Vienna from The Guardian yesterday.

    “Up until recently it was thought that only humans had a permanently descended larynx. Furthermore, it was thought that a lowered larynx represented a pre-adaptation for human speech production, allowing humans to create a wider range of vocal-tract shapes and more varied speech sounds.

    However, we now know that (as well as Koalas) some deer species, the roaring cats, and two species of gazelle also have a descended larynx, indicating that a low larynx position must have alternative, non-linguistic functions. It has been suggested that a descended larynx may be the result of evolutionary pressures to exaggerate size, allowing individuals to gain advantages during sexual competition.

    By revealing this interesting adaptation in koalas, we have shown that vocal adaptations allowing callers to exaggerate (or maximise) the acoustic impression of their size have evolved independently in marsupials and placental mammals.”

  7. AR

    Angra – of course breath control is not sufficient. Dolphins, long before humans, mastered vocalisation and have shown they can easily mimic passable english but, thus far, not grammar.
    I’m a great fan of Chomsky politically but find that his theory of hard wired language begs too many questions and pre-suppostions.
    I specified mammals because of the ‘apparent’ vocalisation of birds. Like the non aquatic mammals their noises are more like ‘throat singing’ than using a larynx and has almost nothing to do with communication of a message, apart from ‘here I am’. Far more information is conveyed by posture, behaviour & olfactory signals.
    Communication, whether bee dances or dog barks, are not speech which I thought was the topic.
    I’d be more than a little wary of invoking Lorenz, Godwin an’ all.

  8. Beware of scientists bearing fraud | Fully (sic)

    […] Crikey's Language Blog Skip to content « Beware of speech experts bearing science […]

  9. Angra

    So here’s the philosophical question.

    Could a bee lie?

  10. Angra

    Hey AR – I forgot bees and von Frisch.

    Karl Ritter von Frisch (20 November 1886 – 12 June 1982) was an Austrian ethologist who received the Nobel Prize in Physiology or Medicine in 1973, along with Nikolaas Tinbergen and Konrad Lorenz.

    His work centered on investigations of the sensory perceptions of the honey bee and he was one of the first to translate the meaning of the waggle dance. His theory was disputed by other scientists and greeted with skepticism at the time. Only recently was it definitively proved to be an accurate theoretical analysis.

    Bees can talk! (maybe)

  11. Angra

    If anyone’s interest in taking this further here are some refs.

    – Wittgenstein – “Philosophical Investigation” – the private language argument.
    – Cassirer – “Philosophy of Symbolic Forms”
    – Langer – “Philosophy in a New Key”; “Feeling and Form: A Theory of Art”; “Language and Myth” (translation of Cassirer).
    – Searle – “Speech Acts”; “The Rediscovery of the Mind”; “The Construction of Social Reality”
    – Vygotsky – “Thought and Language”
    – Bennett – “Linguistic Behaviour”
    – Wisdom – “Other Minds” (also read his “Parable of the Invisible Gardener”)

    My excuse – an aborted research project interrupted by a messy divorce and family tragedy.

    Maybe someone else could take it on.

    I’m too old and tired.

  12. Angra

    Also for some background, Searle’s point goes back the Wittgenstein’s argument ‘Can their be a private (ie. and individually-developed) language?

    His answer is No.

    It’s necessarily a social thing.

    Both Wittgenstein and Cassirer take inspiration from Immanuel Kant, and probably beyond that back to Plato.

    So there’s nothing much new in the world.

    (PS. also the much maligned Chomsky made his name by investigating the notion of ‘depth grammar’ across all human languages. Similar historical roots to the above.)

  13. Angra

    AR – interesting point. I suggest mammalian motor control is a necessary but not sufficient condition for speech. Also what about dolphins and parrots?

    Also this doesn’t consider the importance of imagination, thought and mental imagery which are I think are necessary for a true symbolic language to develop, as opposed to meaningful sounds which can repetitively reinforce a particular response to eg. the sighting of a predator.

    My dog can respond intelligently to maybe a dozen spoken sounds, but this doesn’t mean he can carry on a conversation.

    Again I think Cassirer had it nailed.

    This parallels the debates in artificial intelligence about whether a computer can truly learn a language – eg. the whole John Searle debate about what is meaningful as opposed just accurate repetition of sounds (the Chinese translator machine mental experiment).

    (disclaimer – I am coming at this from the philosophical angle)

  14. AR

    I would suggest that the primary physical skill needed for articulation, leading to singing or speech is breath control. Although many primates can ‘sing’ the sounds aren’t complex.
    The only mammals with sufficient control over breath release, to vibrate those vocal cords, are the aquatic mammals with a diving reflex.

  15. Angra

    What this also ignores is the importance of the development of mental constructs and thought processes in parallel with language.

    A bloke called Ernst Cassirer wrote three epic books about this in the 1920’s – The Philosophy of Symbolic Forms.

    “Man, says Cassirer…is a “symbolic animal”. Whereas animals perceive their world by instincts and direct sensory perception, man has created his own universe of symbolic meaning that structures and shapes his perception of reality – and only thus, for instance, can conceive of utopias and therefore progress in the form of shared human culture.”

    His student Suzanne Langer, went further and applied this to music and other artistic pursuits. Langer believed symbolism is the central concern of philosophy because it underlies all human knowing and understanding. And thus also language.

    Without symbols, there can be no language. And our earliest examples of symbolism are found in myth, religion, dance, music and painting.

    So maybe we had to learn to dance before we could speak?

  16. Rosey Billington

    I certainly had coniptions of disbelief myself when I heard what Frenkel had to say. I particularly enjoyed it when, following his claim that “we stood up and our voiceboxes dropped”, Frenkel suggested that Philip Adams try talking on all fours to see just how difficult it is to talk when you’re not bipedal – and Philip Adams actually gets down on all fours and agrees that it is, indeed, difficult to speak in that position. No it’s not, Philip Adams. That’s just silly, and it makes you complicit in Frenkel’s misinformation.
    Aidan, I also noticed that Frenkel seems to have separate notions of ‘speech’ and ‘language’ which are not clearly defined, but he did reveal some of his concept of ‘speech’ when he said “one of the misconceptions about speech is that it’s all about delivery – but it’s also about content”. Clearly, what he is talking about here is ‘public speaking’, not ‘speech’, and I think that Frenkel confuses these as one and the same thing.
    I have no issue with someone analysing a politician’s skills (or lack of skills) in ‘public speaking’, but passing this off as ‘voice analysis’ and presenting your observations as not just empirically based fact, but also as revealing some inherent characteristic of the politician, is extremely disingenuous. In reality, Frenkel’s analyses are just subjective judgements based on how well public speaking skills relate to qualities that he holds important in singing.
    Frenkel’s ideas are so completely uninformed by all the things that linguists, phoneticians, and speech pathologists have worked hard to learn about language that it is really quite a worry that he is out there promoting himself as an expert on the evolution and analysis of speech.
    The Age, please, next time call in the phoneticians.

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details