Last week, I traveled to Bournemouth to give a talk at the Arts University. I think I got lucky with the weather, and it was a pleasure to enjoy the mild temperatures, intermittent sunshine and bouts of dryness. The other pleasure was addressing Bournemouth’s staff and students. They managed to sit still for a full hour while I pontificated about ventriloquism and cinema. This was the first outing for some new research I’m working on, drawn from a bigger (and long-gestating, oft-delayed) project on Cinema and Puppetry. It’s coming along slowly, but it’s getting there and gathering some speed now that I have more time to devote to it. AUB’s Animation Research Pipeline talks (of which this was one) provides a space for people like me to share work in progress.
I made a complete screen recording of my talk, and while my voice is quite clearly recorded, some of the sound on the clips might need you to raise the volume once or twice. I hope you enjoy it, but I’d love to hear any comments you have, good or bad; it’s not a short lecture, and the first half is quite theoretical, but I promise you it contains good stuff on Charlie McCarthy, The Great Gabbo, Lon Chaney in drag, Mel Gibson having a fight with a glove-puppet beaver, and tastefully coloured Keynote slides.
Here’s the video. It’s available in HD, which should help you if you want to read the text on the slides:
- If you’d like to hear more of the wonderful Edgar Bergen and Charlie McCarthy radio shows, there are plenty of episodes freely available online, especially at the Internet Archive. I have previously posted on this site their 1936 short film Nut Guilty, which is well worth ten minutes of your time. I refer in the talk to a saucy exchange between Charlie and Mae West: you can hear part of it here.
- You can found out much more about Steven Connor’s superb book Dumbstruck: A Cultural History of Ventriloquism here. The book’s website includes many links to further information and articles.
Spectacular Attractions 
According to uncanny valley theory, the dummies less-than-real appearance is not necessarily to maintain room for a commentary on ontological identity (ie. doesn’t exclude it) but can also be due to a more realistic dummy becoming (counter-intuitively) less effective in terms of maintaining the illusion (or otherwise) of a subjectivity ontologically independant of the ventriloquist. There are a few theories as to why this might be case, one being to do with an otherwise sensible aversion to corpses, ie. if the dummy looks too much like a real person it could also look too much like a corpse (or zombie), and we might find it harder to suspend disbelief or identify with it (or whatever we’re doing with it). Of interest is that this aversion is found to evaporate in relation to “photographically constituted people” (as one way of putting it). For whatever reason we are not inclined to recoil from “photo-people” as if they might be zombies, even though (as apparitions, illusions etc) one could argue they at the pointy end of life-likeness, ie. where aversion might have otherwise been predicted at a maximum. The valleys of uncanny valley theory is that chasm, and perhaps a bottomless void, between that which varies smoothly in terms of life-likeness (from muppets to zombies) with increasing aversion, and that which, on the other side of the valley, somehow escapes this.
Re-reading my post I realise an error. The aversion doesn’t kick in straight away. As something approaches more life-likeness the aversion decreases (the curve rises), but at some threshold, it changes direction, and the aversion increases (the curve drops – down into the proverbial valley). It then rises, suddenly, out of the valley, in relation to photographically constituted people. There is a “sweet spot” (no doubt culturally and historically variable) where the life-likeness is not too life-like as to fall into the valley (becomes aversive) but life-like enough to entice a maximum illusion. The use of Bergenesque dummies in horror film dummies may very well be exploiting some historical variability with respect to this sweetspot. What might have otherwise read as a lovable McCarthy in it’s day can read as slightly creepy today. Perhaps in the future, Kermit the frog will read as creepy.
One way of interpreting an uncanny valley curve, is that low-fi dummies are appealing precisely because, as suggested in the talk, they do allow ontological play, if only the possibility of such. Perhaps we have an aversion to corpses because they are not capable of such play. However zombies are different story. They invite aversion, but like puppets, they are very much contributors in ontological plays. So uncanny valley theory isn’t really a critique of the ventriloquial theory being workshopped here, but more of a proposed check on possible pitfalls in it’s elaboration. Play is definitely at work. There is a profusion of ample evidence in this regard. Quite exquisite and beautiful plays. The pitfalls are in terms of how such might be re-projected onto photographic people (I want to say real people), as if they too must necessarily operate in the same way – as ventriloquised.
lots of food for thought. Thanks Dan. Great clips!