WEBVTT - From the Vault: The Kuleshov Effect, Part 2 0:00:05.760 --> 0:00:07.320 Hey, are you welcome to Stuff to Blow your Mind? 0:00:07.320 --> 0:00:09.960 My name is Robert Lamb and I'm Joe McCormick, and 0:00:10.000 --> 0:00:13.200 it's Saturday. Let's head into the vault this week. It 0:00:13.560 --> 0:00:16.200 is part two of our series on the Coolest Jov Effect. 0:00:16.320 --> 0:00:25.040 This episode originally aired January. I hope you enjoy Welcome 0:00:25.120 --> 0:00:34.760 to Stuff to Blow Your Mind production of My Heart Radio. Hey, 0:00:34.920 --> 0:00:36.920 welcome to Stuff to Blow your Mind. My name is 0:00:37.000 --> 0:00:39.920 Robert Lamb and I'm Joe McCormick, and we're back with 0:00:40.000 --> 0:00:43.200 part two of our series on the Coolest Jov Effect. Now, 0:00:43.240 --> 0:00:46.040 as I explained last time, this is one that originally 0:00:46.080 --> 0:00:48.040 was going to be one episode. We ended up splitting 0:00:48.080 --> 0:00:51.480 it into so we're doing a little time traveling right now. 0:00:51.520 --> 0:00:54.400 This is an out of sequence introduction, but I guess 0:00:54.400 --> 0:00:56.720 from here we'll just jump right back into the middle 0:00:56.760 --> 0:01:00.320 of our conversation from last time. Let's do it well. Anyway, 0:01:00.400 --> 0:01:03.040 So I wanted to talk about a very interesting paper 0:01:03.520 --> 0:01:07.280 that analyzed the the history and meaning of the cool 0:01:07.319 --> 0:01:12.080 Shov effect and then also tried to recreate the Mojukan experiment. 0:01:13.120 --> 0:01:16.960 So this paper was published in the Cinema Journal by 0:01:17.400 --> 0:01:21.559 Stephen Prince and Wayne E. Hensley called the Coolershov Effect 0:01:21.680 --> 0:01:26.360 Recreating the classic experiment your nineteen. I think both of 0:01:26.400 --> 0:01:30.399 the authors on this paper were at the time professors 0:01:30.440 --> 0:01:34.280 at Virginia Tech. Stephen Prince is a is a film 0:01:34.319 --> 0:01:35.800 scholar who I know has done a lot of work 0:01:35.800 --> 0:01:39.080 on a Kirakurasawa. And I'm not going to cover the 0:01:39.200 --> 0:01:41.480 entire paper, but i just want to note some parts 0:01:41.480 --> 0:01:44.120 of it that struck me as as relevant and interesting. 0:01:44.280 --> 0:01:47.880 So they start off by telling the story of the 0:01:47.920 --> 0:01:51.680 Kolashov effect experiment, the experiment with that actor mojuk In 0:01:51.760 --> 0:01:54.400 making the neutral face and then either being intercut with 0:01:54.400 --> 0:01:57.280 with soup or with the woman in the coffin, and 0:01:57.400 --> 0:02:01.040 the audience is raving about how how expressive and powerful 0:02:01.120 --> 0:02:04.560 the emotions in the performance were. Now, one thing they 0:02:04.560 --> 0:02:06.760 do at the beginning is they note some differences in 0:02:06.800 --> 0:02:10.239 the details of the story that arise from different recountings 0:02:10.240 --> 0:02:12.880 of it, and so they end up casting doubt on 0:02:13.680 --> 0:02:17.000 whether the accounts of this experiment are first of all, 0:02:17.080 --> 0:02:21.720 historically accurate and second analytically valid, and so the author's 0:02:21.840 --> 0:02:24.320 right quote. The goal here is to provide a clearer 0:02:24.360 --> 0:02:29.560 contextualization of Kolashev's work distinguishing between its incontrovertible importance for 0:02:29.600 --> 0:02:33.919 an understanding of how cinema communicates and certain of its limitations, 0:02:34.040 --> 0:02:39.360 especially it's incautious merging of theoretical claim and observational assertion. 0:02:39.760 --> 0:02:42.440 As we will see, Kulashev may have been right, but 0:02:42.560 --> 0:02:45.960 perhaps for the wrong reasons. So the top line of 0:02:45.960 --> 0:02:49.359 this paper is that they try to recreate the Majukan 0:02:49.440 --> 0:02:52.000 experiment as it is usually described, and they do not 0:02:52.200 --> 0:02:56.200 produce the same result. But this doesn't necessarily mean that 0:02:56.240 --> 0:03:00.800 the broader implications of the Kolashov effect are wrong theoretically, uh, 0:03:00.840 --> 0:03:03.320 but it might mean something about the specific claims about 0:03:03.320 --> 0:03:08.000 a neutral face. Um. So they start off talking Aboutkulashev's 0:03:08.080 --> 0:03:11.560 belief in the power of montage and his arguments that 0:03:11.760 --> 0:03:14.680 editing is far more important the meaning generated by a 0:03:14.720 --> 0:03:18.119 film than the contents of the shots. So they talked 0:03:18.120 --> 0:03:20.760 about the masou Can experiment and then the other things 0:03:20.760 --> 0:03:25.160 we mentioned, creative geography and creative anatomy, and they described 0:03:25.200 --> 0:03:29.480 the general takeaway from the masuk And experiment as follows. Quote. 0:03:30.080 --> 0:03:35.480 Naturalistic emotive performances by actors were not considered by Kulashev 0:03:35.520 --> 0:03:39.800 to be essential to cinema because of the demands of montage, 0:03:39.960 --> 0:03:44.840 actors were to provide minimal restraint and fairly unambiguous gestural 0:03:44.920 --> 0:03:48.680 and facial expressions. As kola Chev puts it, quote, the 0:03:48.720 --> 0:03:54.760 presence of montage necessitated that the shots should be constructed simply, clearly, distinctly. 0:03:55.320 --> 0:03:58.440 Otherwise the flickering of a rapid montage would not be 0:03:58.480 --> 0:04:01.920 sufficient for a full scrutin any of its contents. And 0:04:01.960 --> 0:04:05.200 then the authors go on reacting, partly against the over 0:04:05.280 --> 0:04:09.200 emoting found in some silent films. Kulashov noted that quote, 0:04:09.200 --> 0:04:14.120 a preoccupation with psychologism rooted in the actor's performance was 0:04:14.400 --> 0:04:17.680 quite useless for the cinema. So in a in a 0:04:17.680 --> 0:04:19.679 lot of ways, it sounds like Kolashov kind of wanted 0:04:19.720 --> 0:04:22.599 to take the acting out of acting. He was like, 0:04:22.839 --> 0:04:26.080 there's too much psychology and acting. What we need instead 0:04:26.320 --> 0:04:29.840 is just sort of like shots of actors doing kind 0:04:29.839 --> 0:04:33.680 of like plain, unambiguous moments that can then be selected 0:04:33.720 --> 0:04:37.160 by the editor to insert in a sequence to make 0:04:37.240 --> 0:04:40.839 meaning of m Yeah, that's I mean, it reminds me 0:04:40.839 --> 0:04:45.920 of so many other discussions we've had about performance and direction. Uh. 0:04:46.360 --> 0:04:49.480 I'm always reminded of that that final sequence from A 0:04:49.560 --> 0:04:53.359 Geary The Wrath of God, where you have you have 0:04:53.440 --> 0:04:56.520 what ends up being a rather balanced and and and 0:04:56.720 --> 0:05:00.720 interesting performance by klaus Kinsky. But apparently it's because Vernon 0:05:00.760 --> 0:05:03.760 Herzag just wore him out, made him do take after 0:05:03.839 --> 0:05:07.880 take until he wasn't doing like a frenzied um uh 0:05:07.920 --> 0:05:12.040 you know, over almost you know, overacting overly intense performance. 0:05:12.560 --> 0:05:15.000 He's not raging, he's just actually you know, you know, 0:05:15.920 --> 0:05:18.120 emoting it at the level that the director wants and 0:05:18.120 --> 0:05:22.400 then can therefore be uh be used effectively in the edit. Yeah, 0:05:22.839 --> 0:05:24.560 and though if that story is true, it may have 0:05:24.600 --> 0:05:26.960 worked in this case. So I want to say I 0:05:27.000 --> 0:05:32.039 do not necessarily endorse directing by exhaustion. No. Now that 0:05:32.160 --> 0:05:35.599 was a special relationship obviously, But you often see this 0:05:35.640 --> 0:05:37.720 brought up, and you know, there's this idea of like, 0:05:37.880 --> 0:05:40.919 is this is it the is is this about the 0:05:40.960 --> 0:05:44.920 actor and the acting performance? Is it about uh editing? 0:05:45.000 --> 0:05:48.080 Is it about the director's vision? And you do often 0:05:48.080 --> 0:05:49.960 see that sort of push and pull be a you know, 0:05:50.080 --> 0:05:54.120 Klauskinski in vern Or Herzog or Jimmy Stewart and Alfred Hitchcock. Uh, 0:05:54.160 --> 0:05:56.479 you know, the the actor has a certain vision about 0:05:56.480 --> 0:05:58.960 how things want need to be, and then the director 0:05:59.000 --> 0:06:01.839 had maybe has another eye dea not not only about 0:06:01.880 --> 0:06:04.640 like this particular character in this particular performance, but how 0:06:04.720 --> 0:06:07.960 it fits into the the overall film, how it fits 0:06:07.960 --> 0:06:10.480 into the final edit. And so you could you can 0:06:10.560 --> 0:06:13.000 imagine somebody going into it with this this sort of 0:06:13.080 --> 0:06:16.159 very kulashav idea of just shoot, all we needed as 0:06:16.240 --> 0:06:19.640 neutral actors, we don't really need any of this emotion 0:06:19.680 --> 0:06:22.159 one way or another. And um, I don't know, there's 0:06:22.160 --> 0:06:24.880 probably some examples of filmmakers who tend to lean in 0:06:24.920 --> 0:06:29.559 that direction with very neutral performances. Yeah, you could almost 0:06:29.600 --> 0:06:32.200 look at that approach as uh, something that might be 0:06:32.240 --> 0:06:35.440 more common saying like music videos and stuff than in 0:06:35.560 --> 0:06:38.520 narrative films. Being probably find at some in narrative films 0:06:38.520 --> 0:06:43.160 as well, where the filming part of the filmmaking process 0:06:43.440 --> 0:06:46.000 is just sort of like creating a bunch of building 0:06:46.040 --> 0:06:50.560 blocks that can later be used in various arrangements to 0:06:50.640 --> 0:06:55.359 do whatever the director editor later decides to do with them. Yeah, 0:06:55.520 --> 0:06:58.200 it also reminds me how, you know, a lot of 0:06:58.240 --> 0:07:00.760 the films we're watching weird how cinema will sometimes feature 0:07:01.520 --> 0:07:05.120 non actors or you know, very very green actors, But 0:07:05.960 --> 0:07:09.200 the right sort of non actor can really excel in 0:07:09.240 --> 0:07:11.560 a scene if utilized correctly, you know, like not the 0:07:11.640 --> 0:07:14.160 kind of non actor where they're just really outrageous with 0:07:14.800 --> 0:07:17.720 you know, but but where they're just sort of very 0:07:17.920 --> 0:07:20.200 they're very neutral, they're they're almost barely there at all, 0:07:20.680 --> 0:07:22.520 and if enough of the other stuff is in the 0:07:22.600 --> 0:07:25.080 right place, it can really work. Now. I gotta say, though, 0:07:25.120 --> 0:07:28.560 as this paper ends up describing kola Chev's theory of 0:07:28.720 --> 0:07:32.720 film and montage, I think I can't agree with with 0:07:32.760 --> 0:07:35.320 what it sounds like Coolshov's vision actually was, because cool 0:07:35.320 --> 0:07:38.640 a Show apparently said things like the film shot is 0:07:38.680 --> 0:07:42.120 not a still photograph. The shot is a sign a 0:07:42.320 --> 0:07:46.320 letter for montage. So I think he's saying like a 0:07:46.360 --> 0:07:49.360 still photograph can have meaning on its own, but a 0:07:49.400 --> 0:07:52.560 shot in a movie is more like a letter in 0:07:52.720 --> 0:07:56.160 a sentence. Something which does not have meaning on its own, 0:07:56.560 --> 0:08:00.320 but is combined in sequence to make meaning clearly has 0:08:00.360 --> 0:08:03.040 some truth to it, because, as we've said, editing does 0:08:03.280 --> 0:08:06.080 constitute a major part of the the sense making or 0:08:06.120 --> 0:08:08.280 meaning making of a film. But I think that's also 0:08:08.320 --> 0:08:11.120 pretty overstated. You know, a lot of meaning lies in 0:08:11.160 --> 0:08:13.800 the editing, but the contents of the shots also stand 0:08:13.800 --> 0:08:17.240 alone to a greater extent and and matter a lot 0:08:17.320 --> 0:08:21.400 more than Kolashaw was giving credit here. Um. Though, again, 0:08:21.440 --> 0:08:23.400 to be fair, I think it's important for us not 0:08:23.480 --> 0:08:26.600 to forget that in the nineteen teens and early nineteen twenties, 0:08:27.240 --> 0:08:30.440 you know, film was still fairly young, Editing was still 0:08:30.480 --> 0:08:34.760 fairly new in cinema, and its powers were still being discovered. Uh. 0:08:34.800 --> 0:08:37.040 You know, it's like, like we talked about the very 0:08:37.120 --> 0:08:40.439 earliest films from the eighteen nineties and such, we're usually 0:08:40.480 --> 0:08:43.199 not edited at all. They'd just be one continuous shot. 0:08:43.720 --> 0:08:46.960 And even after editing was introduced, films of the Silent 0:08:47.040 --> 0:08:50.200 era typically did not have as many cuts as movies 0:08:50.200 --> 0:08:54.120 were used to today. Furthermore, the authors of of this 0:08:54.720 --> 0:08:58.640 paper argue that a theory comparing film to language is 0:08:58.679 --> 0:09:01.200 actually not super useful because there's just a lot of 0:09:01.200 --> 0:09:04.120 ways in which that doesn't work. Like film does things 0:09:04.200 --> 0:09:07.480 language cannot do. So you don't have to learn a 0:09:07.559 --> 0:09:10.400 language to appreciate the meanings of films. You you learn 0:09:10.520 --> 0:09:13.040 some conventions, but you know, you can just watch a 0:09:13.080 --> 0:09:15.040 movie and make some sense of it, even if you're 0:09:15.040 --> 0:09:18.480 not familiar with conventions. You to understand the language, you 0:09:18.520 --> 0:09:22.520 have to learn the language. Um. Meanwhile, language does things 0:09:22.520 --> 0:09:25.600 that film can't do, like photographic images used in a 0:09:25.679 --> 0:09:29.480 film cannot be recombined freely to make endless meaning the 0:09:29.480 --> 0:09:33.280 way a language can. There's also an interesting digression in 0:09:33.280 --> 0:09:37.160 this paper about Kolashaw being influenced by the ideology of 0:09:37.240 --> 0:09:42.160 industrial efficiency on the model of the American engineer Frederick Taylor, 0:09:42.280 --> 0:09:46.040 who was a big proponent of finding ways to make 0:09:46.080 --> 0:09:49.640 you know, production processes and factories more efficient, finding all 0:09:49.679 --> 0:09:53.360 the places where waste and and and problems creep in 0:09:53.440 --> 0:09:58.679 and eliminating those, And that Taylor's ideas of industrial efficiency 0:09:58.679 --> 0:10:01.439 were apparently very popular in the Soviet Union at the time, 0:10:02.080 --> 0:10:05.240 and that in a way, the authors say that you 0:10:05.280 --> 0:10:10.480 could view Kolashaw's emphasis on economy and acting as a 0:10:10.600 --> 0:10:15.080 type of industrial efficiency technique applied to film theory. Yeah, 0:10:15.080 --> 0:10:17.160 And based on what I was reading it, it doesn't. 0:10:17.360 --> 0:10:18.920 It does seem like a lot of his work was 0:10:18.960 --> 0:10:22.080 based in let's figure out what's working, and then how 0:10:22.120 --> 0:10:24.480 we can we can do that? How how do we 0:10:24.559 --> 0:10:27.760 make how do what is the most economic means of 0:10:27.800 --> 0:10:31.680 making effective film now? Ultimately, Prince and Hensley make the 0:10:31.720 --> 0:10:35.560 case that Koulishev really was trying to dress up his 0:10:35.720 --> 0:10:40.360 theoretical convictions about how film works with the imper mater 0:10:40.480 --> 0:10:43.520 of empirical science with this alleged experiment them as you 0:10:43.600 --> 0:10:47.079 can experiment, uh, And I think I'm pretty convinced by 0:10:47.200 --> 0:10:49.280 their description of it that way. I think this is 0:10:49.320 --> 0:10:51.880 something you've always got to be cautious of because obviously, 0:10:52.480 --> 0:10:56.280 you know, I don't object in principle to exploring or 0:10:56.320 --> 0:11:00.480 building upon artistic theories with empirical methods. But I would 0:11:00.480 --> 0:11:02.800 also say, my personal opinion is that a lot of 0:11:02.840 --> 0:11:08.320 these efforts to inject scientific methods into esthetics and and 0:11:08.520 --> 0:11:12.640 art and stuff can be confusing and unnecessary. Like I 0:11:12.679 --> 0:11:16.679 don't think you have to have an empirical scientific justification 0:11:17.080 --> 0:11:19.959 for an opinion about where meaning comes from in art 0:11:20.080 --> 0:11:23.360 or in film. Obviously, I'm a huge believer in empirical science. 0:11:23.679 --> 0:11:26.559 I just don't think it has to pervade every domain, 0:11:26.760 --> 0:11:31.040 Like aesthetics and art don't necessarily need scientific evidence and 0:11:31.120 --> 0:11:34.240 theories behind them that those fields just you know, work 0:11:34.320 --> 0:11:36.680 by different standards. And I think also a lot of 0:11:36.679 --> 0:11:42.600 times if you try to generate empirical scientific justifications for 0:11:42.679 --> 0:11:46.680 your beliefs about art or aesthetics or whatever, you're often 0:11:46.720 --> 0:11:50.760 just gonna end up doing sloppy experiments or drawing unjustified conclusions, 0:11:50.760 --> 0:11:54.080 even if you do a good one. Yeah. Um, like 0:11:54.320 --> 0:11:56.360 I'm reminded, you know of the fact that obviously you 0:11:56.400 --> 0:11:58.960 have a such thing. There's such a thing as outsider 0:11:59.080 --> 0:12:04.040 art and oursider cinema. Um, and and examples of outsider 0:12:04.120 --> 0:12:06.800 art and outsider cinema can be amazing, uh, you know. 0:12:07.080 --> 0:12:10.040 And on the other side of things, you don't hear 0:12:10.080 --> 0:12:16.600 as much about maybe outsider architecture, outsider structural engineering, things 0:12:16.600 --> 0:12:20.640 of this nature. Outsider medicine is probably you know, best 0:12:20.880 --> 0:12:23.760 avoided if you can, no matter how it's being dressed up. Well, 0:12:23.760 --> 0:12:26.880 I mean, I think empirical methods are good for fields 0:12:26.880 --> 0:12:31.360 in which you are trying to achieve very clearly specified goals, 0:12:31.559 --> 0:12:35.240 certain kinds of outcomes and get them as reliably as possible. 0:12:35.720 --> 0:12:39.679 And empirical methods are are less important in fields where 0:12:39.720 --> 0:12:43.000 you're you're just trying to be expressive or be creative 0:12:43.000 --> 0:12:46.560 and see what kind of emergent results come out. But 0:12:46.600 --> 0:12:49.480 if it's like like this turns my mind to like 0:12:49.520 --> 0:12:54.840 a b testing and focus groups used in film and television. Um. 0:12:54.880 --> 0:12:57.200 You know, not not necessarily a bad idea at all, 0:12:57.280 --> 0:12:59.800 especially when you're dealing again with a very mainstream product 0:12:59.840 --> 0:13:02.440 you want to appeal to a you know, a wide 0:13:02.440 --> 0:13:06.040 population of individuals. Um. But you know, there are plenty 0:13:06.080 --> 0:13:07.960 of arguments to be made about it as a potential, 0:13:08.280 --> 0:13:11.160 you know, sloppy experiment. As you say, perhaps one of 0:13:11.200 --> 0:13:14.000 the best critiques of all of this is that that 0:13:14.040 --> 0:13:16.640 episode of The Simpsons, the Itchy and Scratchy and Pucci Show, 0:13:18.480 --> 0:13:22.000 one of my favorites. It's just an old, creaky mirror. 0:13:22.120 --> 0:13:26.079 Sometimes it sounds like it's coughing or talking softly. Yes, 0:13:33.920 --> 0:13:36.920 But anyway, to come back to uh, Prince and Henley's 0:13:36.960 --> 0:13:43.760 description of methodological problems with the common descriptions of Kolashov's 0:13:43.760 --> 0:13:46.720 alleged experiment the Masukan experiment with the neutral face and 0:13:46.760 --> 0:13:48.920 the soup and the and the coffin and stuff. And 0:13:48.960 --> 0:13:51.400 they list a bunch of questions, they say, quote for 0:13:51.440 --> 0:13:54.840 such a seminal and basically uncontested study, there is virtually 0:13:54.840 --> 0:13:59.360 no information available about Kolashov's actual method and procedure. Did he, 0:13:59.480 --> 0:14:02.840 for example, will interview the subjects individually or in a group. 0:14:03.200 --> 0:14:05.520 What did he tell them beforehand about the purpose of 0:14:05.520 --> 0:14:08.400 the presentation, What, if anything, did he tell them about 0:14:08.440 --> 0:14:11.520 the nature of film editing or montage. What was the 0:14:11.559 --> 0:14:14.480 frequency of outlier opinions e g. People who did not 0:14:14.760 --> 0:14:19.360 think Masukan was saddened by the dead woman. Published accounts 0:14:19.360 --> 0:14:23.800 suggest the responses were uniform. Was this so? Unfortunately we 0:14:23.880 --> 0:14:27.000 do not know the answers to any of these questions. So, 0:14:27.360 --> 0:14:31.880 given these limitations, they attempt to recreate and try to 0:14:31.920 --> 0:14:34.920 replicate as best they can the conditions of the original 0:14:34.960 --> 0:14:38.000 experiment to see if they get the same result. So 0:14:38.040 --> 0:14:40.600 what they did was they put together a videotape that 0:14:40.880 --> 0:14:44.880 had some auditions for actors to produce a close up 0:14:44.880 --> 0:14:47.680 shot of a face that was just totally neutral and 0:14:47.760 --> 0:14:49.360 expression list And they had to go through a couple 0:14:49.400 --> 0:14:53.080 of rounds because in the first round the actor's neutral 0:14:53.120 --> 0:14:57.600 face was not perceived as neutral enough by the control group. Um. 0:14:57.800 --> 0:15:00.240 But so so they got a neutral face on a 0:15:00.320 --> 0:15:02.760 video and they did the same thing. They intercut it 0:15:03.240 --> 0:15:06.240 with a woman lying in a coffin, a girl playing 0:15:06.240 --> 0:15:08.240 with a teddy bear and a bowl of soup on 0:15:08.280 --> 0:15:10.800 a table, and they tried as best they could to 0:15:11.320 --> 0:15:15.760 follow Kolashov's cues about what what the cinematography techniques for 0:15:15.880 --> 0:15:17.960 making this work the best would be, so it would 0:15:17.960 --> 0:15:22.120 be UH people visible on a darkened black velvet background. 0:15:22.600 --> 0:15:25.440 Apparently the actors were told that they just needed someone 0:15:25.520 --> 0:15:29.680 to uh to model for an instructional video in which 0:15:29.720 --> 0:15:32.840 they would be required to do an expressionless or neutral face. 0:15:33.640 --> 0:15:36.400 So one difference is that instead of one long sequence 0:15:36.440 --> 0:15:39.280 intercutting with all of them, they did separate sequences for 0:15:39.360 --> 0:15:43.840 each reaction. So for example, it might go face soup, 0:15:44.280 --> 0:15:47.840 face fade out or face coffin, face fade out, and 0:15:47.920 --> 0:15:51.400 each shot was seven seconds long. And the separate sequences 0:15:51.440 --> 0:15:53.520 make sense to me because you might get a different 0:15:53.560 --> 0:15:57.120 reaction with some pairings than you would with others. So 0:15:57.400 --> 0:16:00.160 viewers each saw one sequence selected at random, and they 0:16:00.160 --> 0:16:04.080 were told that the experimenters needed help evaluating an acting performance. 0:16:04.600 --> 0:16:07.040 And then the viewers were supposed to select from a 0:16:07.080 --> 0:16:09.640 list of emotions that they thought were being portrayed by 0:16:09.680 --> 0:16:16.960 the actor. Options included happiness, sadness, anger, fear, Surprise, discussed hunger, 0:16:17.320 --> 0:16:22.000 no emotion, and other Apparently the participants were undergrads at 0:16:22.000 --> 0:16:24.800 a mid Atlantic university. I'm going to assume based on 0:16:24.840 --> 0:16:29.640 the author's affiliations, this was probably Virginia Tech. They said 0:16:29.680 --> 0:16:33.840 that interestingly, film students were excluded from the experiments since 0:16:33.880 --> 0:16:37.640 they might detect the connection to Kolashov and understand what 0:16:37.720 --> 0:16:41.040 the experiment was getting at, which could bias results. And 0:16:41.160 --> 0:16:43.560 in support of this decision, I mean, it seems like 0:16:43.600 --> 0:16:46.040 a good choice either way. But to justify this decision, 0:16:46.440 --> 0:16:50.240 they wrote about another recent attempt to replicate the Monsieur 0:16:50.240 --> 0:16:55.000 can experiment in France among film students who allegedly gave 0:16:55.040 --> 0:16:58.320 answers like the following quote. We know that the man 0:16:58.360 --> 0:17:01.280 does not change his expression, but because of the montage, 0:17:01.360 --> 0:17:04.639 we think we see him change or quote. We know 0:17:04.760 --> 0:17:09.159 the Cooleshov effect and it works. And then Princeton Hensley 0:17:09.160 --> 0:17:11.359 also had a control condition where they showed the face 0:17:11.400 --> 0:17:14.800 to twenty four film students, this time but without any 0:17:14.800 --> 0:17:18.000 inner cutting. They were just showing them the face by 0:17:18.040 --> 0:17:21.560 itself and asking them what emotion it was showing for 0:17:21.680 --> 0:17:24.960 the face that they actually used in the experiment. Percent 0:17:25.080 --> 0:17:27.080 said there was no emotion on the face. So this 0:17:27.200 --> 0:17:29.919 is a very good neutral face. You know that. That 0:17:30.000 --> 0:17:33.440 reminds me though of of use of neutral face uh 0:17:33.480 --> 0:17:37.960 sort of not still pictures, but just seen sequences where um, 0:17:38.000 --> 0:17:41.400 a character and individual is staring directly into the camera. Um. 0:17:41.440 --> 0:17:48.360 I'm thinking it's certainly about Ron fricksm Baraka, which features 0:17:48.400 --> 0:17:51.119 a number of these uh sequences where you'll you'll just 0:17:51.160 --> 0:17:54.000 have an individual from from one culture or another just 0:17:54.080 --> 0:17:56.919 staring into the camera. Or Another example that comes to 0:17:57.000 --> 0:18:00.199 mind is the film The Mission, where at the very 0:18:00.280 --> 0:18:03.399 end of the film there's you who just have several 0:18:03.440 --> 0:18:07.680 beats of one of the primary characters, uh staring into 0:18:07.720 --> 0:18:12.199 the camera and very neutral expression. And of course you 0:18:12.240 --> 0:18:15.600 have the entire film you've just watched to help uh 0:18:15.920 --> 0:18:20.400 inform your idea of what's going through that that character's head. Um. 0:18:20.480 --> 0:18:22.600 But but but still, it's it's a it's a great 0:18:22.680 --> 0:18:25.199 use of neutral expression, Like he doesn't it doesn't look 0:18:25.240 --> 0:18:29.440 particularly sad in that case, but you in you can 0:18:29.480 --> 0:18:32.480 see sadness in the character. You know. Well, yeah, that's 0:18:32.480 --> 0:18:34.640 a good example, but I think it also does raise 0:18:34.760 --> 0:18:37.840 questions about something that's supposed to be sort of outside 0:18:37.920 --> 0:18:43.120 the the standard interpretation of this of this experiment, which 0:18:43.160 --> 0:18:45.119 is like, well, wait, what are the actual contents of 0:18:45.160 --> 0:18:48.000 the face? Maybe that does matter. That's going to come 0:18:48.040 --> 0:18:50.560 up in the author's interpretation of the results they get. 0:18:51.000 --> 0:18:53.239 But so in the actual experiment they did, they had 0:18:53.240 --> 0:18:56.320 a hundred and thirty seven participants, including the control group. 0:18:56.640 --> 0:19:00.760 In the experimental group. In every condition, whether it was soup, coffin, 0:19:00.880 --> 0:19:05.040 or child, the majority of people said there was no emotion. 0:19:05.119 --> 0:19:07.760 So they saw the face that was supposedly neutral, they 0:19:07.760 --> 0:19:10.280 saw it intercut with whatever it was, the soup or 0:19:10.320 --> 0:19:13.320 the coffin, and they said, nope, there is no emotion 0:19:13.400 --> 0:19:16.800 on this face. In the soup condition, sixty eight percent 0:19:16.880 --> 0:19:21.880 selected no emotion. In both the child and the coffin condition, 0:19:22.000 --> 0:19:25.520 sixty one percent said no emotion, and so comparing that 0:19:25.560 --> 0:19:29.480 to the control group, in the control eight percent said 0:19:29.480 --> 0:19:31.760 there was no emotion, and that dropped down to sixty 0:19:31.840 --> 0:19:35.880 eight in the soup and sixty one in the child 0:19:35.960 --> 0:19:38.520 and the coffin. So you could say this is a 0:19:38.600 --> 0:19:42.000 small increase in perceived emotion, though the authors note that 0:19:42.080 --> 0:19:44.080 for the size of the group they tested, it actually 0:19:44.080 --> 0:19:47.200 doesn't reach statistical significance, so it might just be a 0:19:47.280 --> 0:19:51.160 random fluke. Furthermore, in the cases where the viewers picked 0:19:51.160 --> 0:19:55.280 in emotion, it was usually not the expected emotion, so 0:19:55.320 --> 0:19:58.119 it was not happiness for the child and so forth. 0:19:58.560 --> 0:20:02.080 So either way, this experiment find something somewhere between no 0:20:02.280 --> 0:20:06.080 effect and small effect on perceived emotion, which is a 0:20:06.160 --> 0:20:08.760 very far cry either way from Kolashev's reports about the 0:20:08.760 --> 0:20:13.440 audience is unanimous raving about the actor's subtle emotional performances, 0:20:14.200 --> 0:20:16.080 And so the authors say here that you know, in 0:20:16.160 --> 0:20:19.119 less contrary evidence emerges, it seems true to say that 0:20:19.200 --> 0:20:24.320 quote the Kolashov effect as reported no longer exists, even 0:20:24.400 --> 0:20:27.040 if the effect did play a role at one time, 0:20:27.520 --> 0:20:30.320 though emphasis there should be on as reported, because some 0:20:30.359 --> 0:20:34.600 of the broader implications of it probably do still hold true. Now, 0:20:34.920 --> 0:20:38.040 this raises an interesting question. If we assume, for the 0:20:38.040 --> 0:20:41.920 sake of argument, that Kolashev was basically reporting the results 0:20:41.920 --> 0:20:46.480 of his experiment accurately or with only slight exaggeration, what 0:20:46.600 --> 0:20:49.720 could account between the difference. Why did Kolashov get people 0:20:49.840 --> 0:20:53.080 raving about the subtle emotion in the neutral face, but 0:20:53.280 --> 0:20:56.159 that that didn't really happen in a modern experiment? The 0:20:56.200 --> 0:20:58.320 authors offer some ideas here, and I think they're all 0:20:58.440 --> 0:21:02.880 pretty possible via and certainly interesting. So one would be 0:21:03.440 --> 0:21:08.159 changes in audience expectation. You know, audiences today are accustomed 0:21:08.200 --> 0:21:11.840 to highly effective editing techniques that have been perfected over time, 0:21:11.920 --> 0:21:15.360 such as, like I mentioned earlier, the preservation of eyelines 0:21:15.480 --> 0:21:19.760 to enforce continuity of of perspective and reverse shots. Yeah, yeah, 0:21:19.800 --> 0:21:21.520 I think this is this is a big one. And 0:21:21.640 --> 0:21:23.240 I mean it comes down to like some of the 0:21:23.240 --> 0:21:25.920 basics of what we said earlier about how at least 0:21:25.920 --> 0:21:28.000 for many of us and certainly for me, like trying 0:21:28.000 --> 0:21:31.760 to watch an actual cooler show film is very difficult. 0:21:31.800 --> 0:21:35.080 Like it's just film has come has evolved so much 0:21:35.160 --> 0:21:39.080 since then, um, and and the effects are subtle in 0:21:39.119 --> 0:21:42.480 a way that you really the film only has to 0:21:42.520 --> 0:21:45.680 be even halfway competent to really just draw you in 0:21:45.920 --> 0:21:49.360 and create the illusion. Right, So uh so the author's 0:21:49.440 --> 0:21:51.719 right quote. It may be that a modern audience, by 0:21:51.800 --> 0:21:55.560 virtue of increased media exposure relative to cool A. Shov's day, 0:21:55.680 --> 0:21:59.480 has become accustomed to a more systematic and complex set 0:21:59.520 --> 0:22:02.679 of associate creational cues, such as those supplied by the 0:22:02.720 --> 0:22:07.000 continuity system of editing. And is correspondingly less likely to 0:22:07.040 --> 0:22:10.359 respond to a montage sequence that employs a blank face 0:22:10.400 --> 0:22:14.480 and minimal, if any associated cues within shots. So maybe 0:22:14.480 --> 0:22:19.320 the bar for perceiving emotion in films has has gone up, 0:22:19.400 --> 0:22:21.959 you know, it's just harder to do now. And at 0:22:22.000 --> 0:22:26.520 the time the Kolashov did his experiment, allegedly maybe the 0:22:26.600 --> 0:22:30.399 audiences were just we're just more it was easier for 0:22:30.480 --> 0:22:33.320 them to project that emotion now that There could be 0:22:33.359 --> 0:22:35.359 a number of ways to read that. One way is 0:22:35.359 --> 0:22:38.840 is thinking about how much exposure modern audiences have to 0:22:38.960 --> 0:22:43.000 modern editing techniques. Um the other way, I guess, and 0:22:43.280 --> 0:22:46.199 the authors don't really favor this explanation, but they say 0:22:46.240 --> 0:22:48.600 another way of looking at it is naivete on the 0:22:49.040 --> 0:22:51.560 part of the early audiences. There's some kind of projection 0:22:51.600 --> 0:22:55.120 going on, because maybe early film audiences were just so 0:22:55.200 --> 0:22:59.560 bewildered by moving pictures that they almost like hallucinated projections 0:22:59.560 --> 0:23:02.560 of emotion. And uh, the authors don't think this is 0:23:02.600 --> 0:23:05.800 a very good explanation for one thing, because they argue 0:23:05.880 --> 0:23:08.400 that a lot of the stories that are used to 0:23:08.400 --> 0:23:12.159 to illustrate the sort of bewilderment of early film audiences, 0:23:12.240 --> 0:23:14.880 like that, you know, the semi mythological things about the 0:23:14.880 --> 0:23:17.880 audiences running away from the Loomi air train and stuff 0:23:17.880 --> 0:23:20.560 that they say that I mean, there were sort of 0:23:20.600 --> 0:23:24.119 events of this kind, but they have been mythologized in 0:23:24.160 --> 0:23:28.280 a way that over emphasizes how naive early audiences were, 0:23:28.359 --> 0:23:30.639 and that a lot of these kinds of reactions may 0:23:30.640 --> 0:23:33.480