CARTOONS KILL: casualties in animated recreational theater in an objective observational new study of kids’ introduction to loss of life
BMJ 2014; 349 doi: https://doi.org/10.1136/bmj.g7184 (Published 16 December 2014) Cite this as: BMJ 2014;349:g7184All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I'm late to the game, as I've only run into this article now, 2.5 years later. Having written and produced my own ballet versions of Sleeping Beauty and Cinderella, to avoid copyright issues I worked from the earliest possible sources. These were folk tales, collected by The Brothers Grimm as first published in 1812, both of my examples were copied from Charles Perrault's Fairy Tales published in 1697 from which we commonly recognize the subtitle "Tales of Mother Goose." Perrault's audience were the French Aristocracy, not children. The earlier the version the more gruesome. Many of these oral traditions date back to Medieval times; certainly Ring-A-Round-The-Rosy dates directly back to the Black Death.
Your source material is rather limited, in that you use North American Animation, which are all from only a few companies that had the financial resources to produce them. You are basing most of your study on the decisions of Walt Disney, who was just one man. His source material was a German collection of a French collection of common folk tales of that time and location. Their movie company's original stories like Finding Nemo appear based on either earlier material or merely the same common themes that their animations already contained - again dating back to one man's decisions.
The American Film Industry has always been the most successful and therefore their stories are most widely spread and interpreted to other languages. This narrow use of source material is flawed. Japan also seems to follow the same themes, at least in their successfully exported animated films, but their films are all from after WWII with potential American influences or merely selection that follows the 'standards' of the film industry, again American.
A more accurate depiction of story telling for children should include all other regions of the world, even if they don't have resources to produce such high dollar films. What stories were and are told in other parts of the world, for children vs adults, and do they contain the same level of death? As France and Germany were dominated by Christianity when these stories were told, a study of what other religions and cultures choose to tell their children over the same time period through today would be a more accurate study of human traditions of story telling to children. Perhaps the same themes do exist in the rest of the world, but you did not include them. Nor does this include the nearly wiped out cultures of even Europe, such as the "Pagan" (anything Christians conquered, such as Celtic) traditions.
Your study is therefore a major reflection of one man: Walt Disney, and his use of Charles Perrault's selection and interpretation of older European Christian folk tales. While this is useful to tell parents how to screen what their children watch and how to react to the most available animations, it does not reflect the rest of the world nor the human condition. No conclusion can be drawn from the study about the human condition, without inclusion of other cultures.
Competing interests: No competing interests
We would like to thank the authors of the responses to our paper. We were happy to see that the piece has generated so much discussion. However, many respondents seem to have misread the tone of the article, given it appears in the Christmas Edition. In this response, we address two commentaries in particular: those by Joseph Wos and Dr. Howard Jacobs.
Mr Wos seems to have misread or inadvertently misinterpreted several aspects of our study. We hope this response provides some clarity around the main concerns raised:
1) The Butler was one of the comparison films for adults, and we did not state or imply that it was a children’s film
2) We did not include movies about cars, robots, toys, etc., as it is not clear whether ‘death’ is possible for these characters. As a toy, could Woody (the main character in Toy Story) even be killed? This was unclear to us, and therefore we excluded these films.
3) With respect to on-screen deaths, we recognize that this may be confusing to some readers. However, we defined ‘on-screen’ death, a priori, as cases where there was clear on-screen evidence of the death of a character who had been present in the film, even if the director spared us the most gory evidence. So the shooting of Bambi’s mother, implied by her sudden absence from the film following a panicked dash away from oncoming hunters and the sound of gunshots, was counted as onscreen death. Similarly, Gaston plummets from a rooftop in the final scenes of Beauty and the Beast. We do not need to see the instant he hits the ground to recognize a death scene. However, in films with orphaned princess whose parents were absent from the very start of the film, we did not consider the parents’ apparent death as an on-screen death. This a priori definition of on-screen death was applied to both children’s animated films and the comparison films for adults.
We are also sorry that Mr Wos objected to the tone of our piece. Our aim was to use a scientifically valid methodology to investigate an unusual research topic, in keeping with the BMJ’s Christmas theme. Our paper was kept light-hearted in tone, while recognizing the potentially positive and negative aspects of childhood exposure to death (or the concept of death).
Dr. Jacobs suggests that Table 1 shows evidence that onscreen death is more common in the comparison films for adults. It appears that Dr. Jacobs may have misread “no onscreen death” to mean “onscreen death”. 30 of 45 children’s films included an on-screen death (67%), whilst 45 of 90 comparison films included an on-screen death (50%).
Additionally, Dr. Jacobs questions the provenance of the claim that ‘the risk of murder was higher in children’s animated films than in dramatic films’. We used survival analysis and Cox regressions to test differences in survival between the two types of films (using Hazard Ratios as a measure of this difference). As stated in the methods section, the data gathered were the times at which the first death occurred. Since observers stopped collecting data after the first death, we were not able to analyze information on subsequent deaths (for example, some films may have depicted a murder or parental death after the first death, which would not have been captured by these methods). Using these techniques, the risk of this first death being a parental death was 5 times higher in animated versus comparison films, and the risk of the first death being a murder was 2.78 times higher. We recognize that this type of analysis has limitations, however, we chose this technique as it is standard epidemiological practice when studying human participants. Certainly, different statistical techniques are designed to answer different questions and may yield different results.
Dr. Jacobs also objected to the conclusions drawn from the study, arguing that viewing violence in film does not necessarily beget violence among spectators. We do not argue that children watching such films are likely to become violent – rather, we suggest that if children are not adequately prepared to face these themes, they may potentially be confused or frightened. We would agree with Walt Disney’s assertion that these films are, perhaps, more suitable to older children. However, at least anecdotally, many parents may be under the impression that animated films are appropriate for children of all ages, and may be unaware that these films depict death at all. We therefore would caution parents to inform themselves as to the content of these films (for example, by referring to websites such as www.commonsensemedia.org) and decide for themselves when their child is ready to view them.
We certainly agree with Dr. Jacobs that exposure to death on-screen may be a valuable tool for scaffolding children’s understanding of death, which we mention in our discussion section. We also suggest that parents watch such movies with their children, in order to be present if questions arise, and provide valuable conversation about themes of death.
Mila Kingsbury, Ian Colman, and James B. Kirkbride
Competing interests: No competing interests
In the four decades since Milton Shulman suggested that the increasing time that children spent watching violence on TV might have an unfortunate effect on the behaviour of some of them, (1) the evidence supporting his fear has mounted inexorably.
It is an insult to your readers’ intelligence and worldliness to labour this point.
It may be a measure of the authors’ unworldliness that they have chosen to treat their subject as they do.
If children exposed to cartoon or other forms of screened violence for 3-6 hours a day are influenced by it ( why else does the advertising industry spend billions travelling the same route ), so are adults.
And so are criminal adults.
Listen to brave writer Roberto Saviano, living under constant protection from those he has exposed. “Everyone thinks that cinema imitates crime. In actual fact it is crime that imitates cinema.... killing is cool“. (2)
1 Milton Shulman. The Ravenous Eye. Cassell, 1971
2 Roberto Saviano https://www.youtube.com/watch?v=VZcMkBzMHmo
Competing interests: No competing interests
I find some of logic of your study to be very confusing.
Firstly, you excluded "action" and "adventure" films because they are "often also marketed to, and viewed by, young children." I find this statement incredibly difficult to justify. It is of course unfounded to say, "All action and adventure movies are also marketed towards children", and so more care should have been taken in choosing which movies not to analyze. By excluding the action and adventure genres - particularly action - you exclude many movies that are likely to portray death; especially in violent ways.
Secondly, by limiting deaths to important characters as defined by the parameters you chose, you potentially exclude a number of particularly violent deaths that occur to non-important characters. For example, in the movie "Black Hawk Down", there are a total of 135 deaths; 116 of which were "Somali militia", which presumably did not meet the criteria to be considered important characters. (http://moviebodycounts.proboards.com/thread/39/black-hawk-unrated-extended)
By only picking important characters, you dramatically undermine the real number of potentially traumatizing deaths that appear in these movies. It is possible that by only examining movies tagged as "drama", and then by only analyzing deaths amongst the main characters, their friends, families, and antagonists, that you are making unrealistic comparisons between movies intended for adults and movies intended for children. The list of movies that were selected for comparison might not be an accurate representation of the contents and variety of movies meant for adults as a whole, and the presented data might not be an accurate representation of the death and violence portrayed in these movies.
The extent to which the criteria for "comparable adult movies" was refined seems to present a bias.
Furthermore, the study fails to meet its own criteria for considering deaths in certain instances. The death of Tarzan's parents, for example, occurs off scene.
Competing interests: No competing interests
"It is unclear how the inclusion of off-screen death could have influenced our study"
I was rather surprised at this statement. Bambi's mother dies off-screen: her death is strongly implied ("Your mother can't be with you anymore"; "She isn't coming back, is she?") It is not explicitly referenced. Tarzan's parents also die off-screen; their death is implied in a way that might escape younger viewers.
There are other mistakes: a giant octopus/sea-witch is impaled on a ship's bow in "The Little Mermaid" - it is misleading to refer to this as a "stabbing"! No important character dies on-screen (or off-screen!) in "How to Train Your Dragon". Even if they did, I'm not sure that a "dragon attack" would be classed as an "animal attack" by the coroner.
And, as a rule, the dwarves don't seem all that "vengeful" in "Snow White". (Indeed, one can only quantify their grumpiness as 1/7.)
And a note for the peer reviewer. I am rather sure that the on-screen death of Marion Crane should not be compared with the off-screen death of Nemo's mother. Ms Crane's death is shown from numerous camera angles and she fights for her life for at least three minutes. We see her life washed down the plug hole and her last breath. When Nemo's mother dies we hear Marlin call out "Coral!"
(But I suppose the first five minutes of "Finding Nemo" might be traumatic as the shower scene in "Psycho" if one is aquaphobic.)
I assume, however, that this paper was either a Sokal-type hoax or a satire on studies which generate headlines about media violence. It was rather entertaining!
Competing interests: No competing interests
Clearly, BMJ research shows by these responses that us Americans 1) Just don't get high intellectual cheeky British humor and/or 2) We're just a bunch of fluffheads too busy watching Honey Kardashian Boo Boo to "get" long-running inside jokes.
The article that mentioned null groups for quantifying parachute efficacy is still my favorite. I humbly submit future potential topics to include epigenetic inheritance of brain cell deficits of some sort for Americans who watch aformentioned reality shows and/or lack of null group for said study. I'd love to help coauthor.
Keep up the outstanding humor and research,
An American Ambulance Driver
Competing interests: No competing interests
There are two aspects of the paper by Colman et al (December 16th 2014) that cast serious doubt on the validity of their work. The first aspect relates to the data presented, the second to the authors’ speculations concerning what they purport to show.
Concerning the data presented in the paper:
The authors state that ‘Notably the risk of murder was higher in children’s animated films than in dramatic films for adults.’ I cannot find information in the paper that specifically itemises the cases of murder in the films but, calculating Fisher’s Exact Probability, I have been able to compare the total number of films that are said to depict a death on screen in the two types of film (data reproduced from their Table 1):
Number of films with A death on screen No deaths on screen Total
Films for children 15 30 45
Films for adults 45 45 90
Total 60 75 135
Thus 30% of films for children depict a death on screen, compared with 50% of films for adults, a difference which, while only bordering on statistical significance (P = 0.0702 (two tailed) or 0.048 (one tailed)), is in complete and inverse contrast to the authors’ statement about the depiction of deaths from murder.
The authors state that ‘In the present sample, risk of parental death was five times higher in children’s animated films compared with dramatic films for adults.’ Again, calculating Fisher’s Exact Probability, I have compared the numbers of films in the two groups in which parental deaths are said to be depicted (data reproduced from their Table 2)
Number of films with an onscreen Parental death No parental death Total
Films for children 8 37 45
Films for adults 6 83 89
Total 14 120 134
In this comparison, the depiction of a parental death is 2.6 times, not five times, more common in films for children (17.78%) than in films for adults (6.7%), a difference that again only borders on statistical significance (P = 0.0708 (two tailed), 0.0501 (one-tailed )).
In the section headed ‘The Ugly,’ the authors say they have considered only films in which a death is depicted on screen. They then go on to say that they included Bambi in their analysis. Contrary to what many people recollect, the death of Bambi’s mother is not depicted on screen. The point about whether deaths are depicted on screen has also been made by your correspondent Joseph Wos, whose detailed observations also cast serious doubt on the methodological validity of this study.
Figures 1 and 2 in their paper depict survival curves for important characters in the two sets of films. No data points are shown but in both figures the abscissa and the curves extend to close to three hours. I am unaware of any film made for children that has a duration of three hours and, indeed, there very few films made for adults that last that long. I cannot understand what conclusions can be drawn from the demonstration that a death occurs earlier in one kind of film than another, provided the viewer stays to watch the whole film.
I have tried to get access to the appendix of this paper to explore further the data set with respect to the misclassification of the films identified by Wos, particularly in view of Walt Disney’s answer to a question about the intended audience for Snow White and the Seven Dwarfs: ‘Before seven or eight,’ Walt Disney told a reporter in 1938, ‘a child shouldn’t be in a theater at all. But I didn’t make the picture for children. I made it for adults – for the child that exists in all adults’. Unfortunately clicking on the link provided only returns me to the paper itself.
I conclude that the data presented in this paper, their classification and the authors’ analyses of them are seriously flawed.
Turning now to what the authors purport to show, it is widely acknowledged that films and literature (including folk tales) produced for children often contain disturbing life events and fractured families, as set out in my PhD thesis, Animated Enchantment: A Psychoanalytic Exploration of the Enduring Popularity of Disney’s First Feature Films http://animatedenchantment.blogspot.co.uk/.
As Colman and his colleagues acknowledge, children often watch animated films many times. Unfortunately, it does not seem to have occurred to Colman et al to ask why this should be so or whether the spectator might actually derive some benefit from repeated viewing of the types of films they mention.
Fears and anxieties, both conscious and unconscious, that concern attachment and separation from mother, and madness and the death of parents, are often features of the internal experience of developing children. In my thesis, I argue that many of the dystopic films that children like to watch evoke these existential fears and, through the often-repeated re encounter with them, such films can provide spectators with an opportunity to work through their deepest fears. The simplistic notion that a film which shows violence automatically makes its spectators violent is tediously repeated by Colman et al. Do they really believe watching Snow White and the Seven Dwarfs or Bambi makes us delinquent? Today’s reader deserves a more nuanced approach, along with more valid data than those Colman et al have offered in this very disappointing paper.
Competing interests: No competing interests
It would be interesting in the light of this knowledge to assess the psychological effect of Punch and Judy puppet shows. The genre has remained relatively unchanged for well over 300 years and delighted children through the ages. However, it is much reviled by (some) adults as gratuitous violence against women, glorifying domestic violence, anti-authoritarian, and politically incorrect, This has resulted in the show being banned in some places, and drastically watered down in many. In the light of this research, on face value it would appear that Punch helps prepare children for a violent world and to deal with the emotional response to death.
Competing interests: No competing interests
I remember seeing Bambi when I was about eight years old and I was shocked. I was scarred for life.
Competing interests: No competing interests
Re: CARTOONS KILL: casualties in animated recreational theater in an objective observational new study of kids’ introduction to loss of life
Parents should be aware that even a few minutes of viewing a fast-paced television cartoon significantly worsens cognitive abilities and executive function in children.
References
https://www.ncbi.nlm.nih.gov/pubmed/20229267
http://pediatrics.aappublications.org/content/128/4/644.long
Competing interests: No competing interests