Current Contents

Past Issues



Barker, Martin (writing in a personal capacity):

'Response to Stephen Kline'

Particip@tions Volume 1, Issue 2 (May 2004)


Response to Stephen Kline

Stephen Kline’s critique of the Amici Brief is interesting (Participations, Vol. 1, No. 1), and I am glad that Participations has agreed to publish it.  But it is more than anything interesting, to my eyes, because of what it tells about the state of our field, and our ability to communicate with each other about the problems in ‘violence’ research and debates.  At one point in his critique Kline notes that there are some differences among those of us who put our names to the Brief.  He is right, I am sure.  But I attach a quite different significance to that fact.  To me, a very substantial part of the problem we face is how deeply- and long-entrenched the effects tradition is, and how far it has become in consequence a touchstone topic in media discussions and political/public debates.  These have the effect that even those who are most sceptical about its claims can find themselves nervous about how strongly to express counter-positions, and how to state an alternative position.  Indeed, where does one turn to find the tools, and the language, and the critical resources to state that alternative?  It is no surprise, therefore, that there is not an oppositional position.  (I could add that, even if all my colleagues agreed with every word I write below, it would have of course been tactically necessary to place the emphasis of the Brief roughly where it was, simply to be admissible to the Court – there is nothing wrong in principle with tailoring one’s presentation of a position to be able to reach a particular audience.)  But it is because of this, and because I know that my stance on this represents an outside edge of opposition to the entire tradition of effects research, that I write this very much as a personal response (rather than, in any sense, in my capacity as editor of Participations).

I therefore want to elaborate here a position that I feel is slightly buried within the Brief, a position which argues that the problems with the effects tradition are not first and foremost problems with the evidence.  If my reply to Kline comes across at certain points as a little intemperate, perhaps that should be measured against the blithe moves in his reply to treat people like myself as virtually unpaid spokespeople for the media industries, doing their intellectual dirty work, arguing for uncontrolled media commerce.  This sly accusation, contained in his slides between comments on our argument, and claims about the industries’ motives, are of course part of the very ideological terrain which a number of us have been seeking to identify.  We are asked to be either/or (in rather Bush-like fashion). Either we are for ‘responsible controls’, or we are for a ‘free market’.  I am happy to vent my irritation a little at this implied sneer.

At one point in his argument Kline quotes me as calling effects research “daft” – but then fails completely to understand what prompted this assertion.  Let me then set out the bare bones of the critique which I want to make.  It is not just that the research is over-hyped, or that the conclusions arrived at are less sure than its proponents would like to admit, or that there have been methodological problems in how particular researches have been done, or even that research of differing qualities is conflated, that histories of critical evaluation are lost, and different particular conceptualisations of how to measure ‘violence’ have been pressed mercilessly together (although I agree with colleagues who argue these, that all these are true).  No, I have argued and will argue that the central problems with the effects tradition are these: that they are asking a non-question, a meaningless question – but a question whose meaninglessness has been ruthlessly disguised; that the persistence of those involved in the tradition in asking this non-question betrays all the hallmarks of, not just a fading Kuhnian paradigm which has failed in its research imperatives, but rather an ideological quest, comparable to the quest for witches in the Middle Ages.  And I do mean that analogy very seriously.

To put it at its bluntest, there is no ‘researchable entity’ which can be named ‘television violence’ or ‘screen violence’ which can be researched well or badly.  It is the same order of false question as the question whether there are witch-essences within people.  But just as that non-question claimed to be a grounded research tradition, and built houses of conceptual and methodological cards around themselves which it guarded against scrutiny, and then defended itself by charging its opponents with forms of heresy, so the effects tradition accuses critics like myself with being ‘dangerous’. 

This position is hardly recogniseable when related to Kline’s presentation of the supposed nature of the battle between the Amici Brief’s supporters and its opponents.  In fact, Kline’s opening gambit invites us to read the debate as one based primarily in the contrasting perspectives of (experimental-quantitative) American social scientific and (humanistic-qualitative) European culturalist traditions.  So, he writes of our (the people who put their names to the Amici Brief) “opposition to positivism and quantitative methods” – but then suggests an inconsistency in our adoption of methods of critique which depend upon those very traditions.  His use of Raymond Williams to this end turns the debate into one in which people are essentially talking past each other – and therefore wishing for better dialogue.  This misses something to me essential – that two incompatible conceptualisations of the ‘object’ to be investigated inhere in the two traditions.  This is not just a debate about methods, but crucially about the ownership of the questions to be answered.  I grant that some of my colleagues who came together to produce the Brief may not agree with all I now argue.  But that is a problem of the state of our field.  It is so very hard to make the intellectual space for this position that I am not surprised at incomplete agreements.

So, when Kline writes that “the effects of children’s exposure to violence has become one of  the most researched issues  in media studies”, there are at least two distinct ways in which this undeniable fact can be considered.  It can be considered from the point of view, common to so many commentaries, of asking ‘what is the balance of opinion and probabilities?’.  Or it can be considered from the point of view, which I unequivocally subscribe to, of asking: how does it come about that such a very stupid question has been taken to be so unproblematically obvious, that it has generated a funding tradition, and a paradigm of enquiry, that is so hard to dent?

Or consider when he writes: “Unfortunately the Amici’s insistence that unless experiments show that after playing a murderous video game a significant number of children jump up and kick or hit another child, there is no proof of a harmful effect on children’s behavior is equally incompatible with the ‘falsifiability’ principle subsumed in scientific rules of evidence.”  That insists that the only grounds of debate are to be grounds of the empirical strength of the evidence.  I simply don’t accept that.  The crucial grounds are much wider than that.  They are conceptual – I have argued and continue to argue that the term ‘TV violence’ (or its near-equivalents) is fundamentally incoherent.  It has the same status as terms such as ‘phlegmatic disposition’, or ‘artificial foods’.  These are pre-scientific, ill-conceptualised notions.  And my argument is that ‘media violence’ is just another such.  The difference is that it has built around itself such a wall of ‘scientific’ noise, which has sat very well with a large set of interest groups – including, paradoxically, a set of media interest groups (see on this Willard Rowland’s excellent book The Politics of Television Violence) – that the paucity of its central concept has been allowed to go virtually unquestioned.

So, on the approach I wish to take, bits of terminology that Kline is content to take for granted, become arguable ideological constructs.  Take as one example the concept of “heavy viewers”.  Think the difference that is introduced if we call such viewers “experienced” viewers” … or “devoted” viewers … or “routine” viewers.  The term ‘heavy’ carries implications of increased vulnerability, of cumulative influence.  These are completely untested, because un-noticed, implications, and they obtain their valency from the untested framing assumptions that underpin effects theorising. (I have discussed some of these briefly in an essay I wrote in Approaches to Audiences.)

Once question that central concept, and the apparatus of Kline’s argument becomes strangely foggy.  For instance, at one point he writes: “One of the most intellectually paralyzing assumptions of the Amici’s brief is that violence has always been with us throughout history and is so pervasive in our culture that there is nothing we can do about it.”  Not so at all.  Those who occupy anything like my position do not want to do anything about it, because there is not an ‘it’ that can have one thing don’t about it.  I am not going the simple road of citing Aeschylus, or Shakespeare and the functions of violence in ‘high art’, and wanting to segment that off.  I am interested in the cultural significations and importance of forms of violence in many forms of popular literature, media and representations.  A good many years ago I conducted a piece of research on a British comic book, Action, which was driven off the newsagents’ shelves by a moral campaign against its ‘violence’.  What emerged from an investigation both of its textual organisation and of its most devoted viewers, was a picture which saw the violence of the stories as that which enabled them to operate as critical political sources for them.  Interestingly, those who only found the comic ‘violent’ were those who were least involved or interested in it.  (On this, see my Comics: Ideology, Power and the Critics.)

This example, of course, then raises serious problems for one of the other ‘moves’ in Kline’s argument, to introducing the idea of ‘community standards’.  I quote: “Others among the Amici have claimed that a ‘media panic’ about violence has been primed by moral entrepreneurs who are opposed to violence and sexuality in the media on taste grounds (Buckingham 2000, Davies et al. 2001). But this obscures the status of entertainment as commercial popular arts – that is as cultural artefact and commodity circulated in the market to children. Even assuming the law was intended to maintain moral and artistic taste, is there to be no state interest in maintaining community standards in cultural markets?”  There has long been a ‘smoothing’ assumption with American behaviourist social science that it is possible to identify, independently of all political positions, a distinction between pro-social and anti-social behaviours and attitudes, and the notion of ‘community standards’ belongs right there in the middle of that ‘smoothing’.  My devoted readers of Action are a scandal to such a position, because they refuse to accept that there is some simple ‘community’ to which they belong. 

So, I simply refuse to see the debate as over the ‘evidence’ – because the research which produces the ‘evidence’ is fundamentally flawed, conceptually naïve, and ideologically-laden.  It derives its plausibility from a widespread capacity not to ask simple questions about the absurdity of that central concept: ‘media violence’.  Let me take one example, a tiny one, which nonetheless so beautifully exemplifies these problems.  In the British liberal broadsheet newspaper the Guardian (7 January 2004) appeared an article entitled “Seeing is believing”.  In this article, Robert Winston, currently one of the most popular of science popularisers, reported how he had finally become convinced that television violence can cause violence in the young, through mimicking.  His article told the story of a series of television programmes which he had been asked to front, which the programme makers hoped “could provide an extraordinary insight into the relative importance of nature or nurture in shaping how children grew up”.  One programme shows an experiment in which children are watching a TV screen showing something happening next door.  First, they watch as a man cuddles and strokes a life-size rubber doll – and when they go next door, they do the same.  Next, they watch as he batters the doll with a mallet – and when they go next door, this time “each attacks the doll viciously.  One toddler, normally shy and retiring, is completely carried away – his violence continuing even when his mother comes in and tells him to stop.  It is some time before he can be dragged off and calmed down”.

Winston was so impressed, he says, that he has changed his mind – from being a sceptic about screen violence, he now accepts the case – at least to an extent: “Is there really hard scientific evidence that watching television affects how children communicate?  Well, yes – and the evidence grows steadily.”  And he goes on, inevitably, to report on the research of Eron Huesmann and Jeffrey Johnson.

What is wrong here?  There are very many things, and I want to state the main ones, one at a time.

First, of course, the wrong-ness is that a professor of fertility studies can here become an ‘expert’ reporting on the security of ‘hard scientific evidence’.  Wrong – yet absolutely typical of the field.  His ignorance of the field was sufficient that he didn’t know that the ‘experiment’ he described was a virtual reproduction of the notorious 1960s Bandura ‘bobo-doll’ experiment – an experiment which successive generations of critics have shredded.  Or, worse, he did know this, but chose to ignore all the critical literature on it …  The problem is precisely that ‘expertise’ is now honorifically gained.  Anyone who makes the ‘right’ noises is an expert.

Second, and more importantly, the wrong-ness is that what this group of children did should be described as ‘mimicking’.  The bizarre thing about this is that these children did far more than mimicking.  They set about the doll with a joyous abandon which in itself demonstrated that something quite other than ‘learning’ was going on here.  But one of the central features of the whole violence research tradition is a total laziness about conceptualisations that supposedly scientific concepts of ‘social learning’, ‘aggression triggering’, and ‘identification with aggressive role-models’ can sit so comfortably with popular notions of ‘copy-catting’ and ‘mimicking’.  What does that sterile redescription of what that ‘shy and retiring’ boy is doing, as he senses that for once he can let rip, as ‘mimicking’ do?  It takes him out of the history of his family and his place within in, his education and all that he has learnt about ‘proper’ behaviour, his class and his relations with friends, his sense of the world as a safe or dangerous place.

Third, and even more importantly, the wrong-ness is that very claim that these children are watching television.  They are not – and for a moment it appears Winston acknowledges this. They are, he says, watching a “TV screen”.  But within a couple of paragraphs this has become identical with “watching television”.  How else do we make sense of Winston’s assertion that “It is worth remembering that, in contemporary Britain, the average three- or four-year-old watches a screen for around five hours each day, and more than 50% of three-year-olds have a TV set in their bedroom”? 

Let us briefly rehearse the sheer variety of kinds of watching of screens that children do – and let us be clear that this is a commentary, at this point, just on the nature of the technologies, not yet even on contents.  They watch broadcast television – with the characteristic that programmes are timed events, with little opportunity for interaction other than parasocial responses (although a good number will already be learning that new technologies now allow you to store favourite programmes, and thus to pause and catch up with a programme if your viewing is interrupted).  They watch interactive television – with the characteristic that you can select how materials are displayed on the screen, which ones are heard, what additional information is provided, and sometimes how you yourself feed back to the programme. They watch streamed video – which comes via a PC, using special connections that you have to have the knowledge how to set up and activate. They watch videos – with the characteristic that you can start and stop them, rewind them, rewatch bits of them (it doesn’t matter whether you do these or not, knowledge of these characteristics is part of them). They watch DVDs – which have, characteristically, chaptered presentations, extras, the opportunities to hear commentaries etc.  They watch screens for purposes of playing games – in which they follow, but simultaneously challenge and try to defeat, sets of programmed rules (on this last, see Steven Poole’s excellent account of the principles of construction of computer games).  And in addition to all these, screens of course now have a range of other meanings.  Children are learning that screens also means video cameras (home-owned, operating in shops, public spaces) – which show a ‘world going by’, including themselves.  This is a sample list.  Each one is different, makes different demands on and offers different possibilities to its users.  At the very least we need to hear some good reasons for supposing that children, when watching a screen, are watching as a “television screen”.  Such is the power of the effects discourse, it seems so unnecessary to ask for this.

Fourth, the wrong-ness is in that singular term ‘television’.  Here is the classic move by Winston – who is here saying something no different than has been said by just about any ‘effects researcher’ you can name:  “By the time they are 18, American children will have seen around 16,000 simulated murders and 200,000 acts of violence on TV”.  In what kinds of programming?  Well, all these depend on content analyses.  And every content analysis I have ever seen has to register that the highest ‘count’ of all kinds of violence comes from cartoons.  This is, of course, the standard ‘move’ that people like myself regularly make – because we want to disclose the sheer fatuousness of these kinds of ‘count’. 

But be that as it may, even ignoring this, such counts depend ontologically on treating as one and the same all the following: news[i], documentaries, satires, docudramas, soap operas, old and new films, cartoons (of all genres), adventure serials, ‘real-life’ reports, simulations, and repetitions, and so on.  It has to presume that people watching a first showing of new material respond in a manner that does not distinguish it from watching (or rewatching) old material.  In other words, the ‘count’ has to presume that ‘violence’ means the same, no matter what the channel, format, age, date, or genre of delivery.  It then has to presume that a process of accumulation stores all these influences in the brain – so that a child who does watch all those 16,000 simulated murders and 200,000 acts of violence is more at risk than someone who hasn’t. This has to be down to some process like ‘desensitisation’ – another of the flabby concepts so typical of this kind of research, because it conflates the ways opposites.  It conflates the ways in which people can learn to manage uncomfortable experiences (such as medical workers learning not to faint at the sight of injuries … which does not make them any the less caring about the hurt people have undergone, mind), or learning to cope with highly stressful situations (for example, herpephobia … but as a result of which people do not have to come to ignore the differences between harmless and poisonous snakes) with the ways in which people learning a genre on television learn how to tell the conventions of staging of, for instance, a murder.

Fifth, I am very struck by a wrong-ness of a different order.  It is a feature of a long tradition of American meta-science to limit debate about the nature of human beings to a choice between ‘nature’ and ‘nurture’ – as though these were obvious, transparent, agreed terms for describing the ‘only two halves’ of possible interest to us.  Really?  ‘Nature’ in fact is parsed as ‘the influence of heredity’ and the preconstitution of our nature by genetic factors (all factorable, all in principle disposable as the influence of ‘this gene’ or ‘that chemical substance’).  And ‘nurture’ is parsed as a collection of ‘isolables’: parenting skills, social deprivation, educational level, and so on.  This ‘world’, which presents itself as the only legitimate one, has no space at all for history (how are these children being brought up within traditions, locate themselves within time and space), belonging (to what communities, real and imagined, do these children belong, and how do they orient to them?), understandings (what skills do children sense they are developing and trying out? Why are they important to them?  What senses of pleasure and satisfaction do they bring to them?).  These and very many other questions – which begin from seeing human beings not as some juncture-point of two external pressures – change the emphasis altogether. Television (and all its variants) is a source of meanings and understandings, emotions and relationships, time out and time in, skills and routines, and senses of belonging and exclusion

The nature-nurture paradigm is an ideological paradigm.  Its roots are strongly within an ideology which I would risk to declare primarily an American one.  Its roots are in debates which emerged in the late nineteenth century, when a science concerned with methods of administration and control emerged in the USA.  And it has only recently struck me how deeply the violence-effects tradition is grounded within, utterly dependent upon, that paradigm.  I have recently begun to explore the history of this, and of the ways in which the concept of ‘violence’ as a unitary phenomenon emerged from this (see my essay forthcoming in Steven Schneider’s (edited) New Hollywood Violence).

So, my counter to Kline’s counter is to say that he has substantially misunderstood what the opposition to his tradition is all about.  Yes, the Amici Brief contains some differences. Yes, it is also in part tactically directed to the needs to the situation.  But no, it is not just about the ‘evidence’ or the ‘methods’.  It is about the conceptual (in)coherence of the entire project which has sought to colonise what methods are permitted, and to control what will count as ‘evidence’.  This isn’t the first such tradition.  Social Darwinism was another such. Eugenics also.  And yes, Soviet Lamarckianism too.  Stephen Kline may refuse this idea – his right entirely – but it won’t do any longer to present people like myself as simply ‘declining to accept the evidence’, or as being virtual stool-pigeons for the media industries.  On the contrary, we are pointing to a deeply-embedded ideological process passing itself off as ‘science’.  It doesn’t surprise me that those on the inside of it can’t see much beyond.  It worries me quite a lot that they still manage to corner so much of the terrain of debate – they still own the ‘obvious’, the first sign of an ideological formation.  If this publication of the debate has only the benefit of posing the depths of the disagreement a little more clearly, it will have been worth it.


Contact (by e-mail): Martin Barker



Barker, Martin, Comics: Ideology Power and the Critics, Manchester: Manchester University Press 1989.

Buckingham, David (2000). After the death of childhood : growing up in the age of electronic media. London, Polity Press. 

Davies, H., David Buckingham and Peter Kelly. (2000). In the Worst Possible Taste: Children, Television and Cultural Value. European Journal of Cultural Studies, 3(1), 5-26.

Dickinson, Roger, Ramaswami Harindranath and Olga Linné (eds.), Approaches to Audiences: A Reader, London: Arnold 1998

Rowland, Willard D, The Politics of Television Violence: Policy Uses of Television Research, Beverley Hills: Sage 1983.

Schneider, Stephen Jay (ed.), New Hollywood Violence, Manchester: Manchester University Press 2004.

[i]   I am curious about that word 'simulated', which might seem to exclude news and documentary.  I know of no content analysis that did actually exclude such real programming.  But if they did, I would want to ask if they also excluded 'police-action' programmes … or historical reconstruction films … or drama documentaries … etc.  It is another example of a completely hollow conceptualisation which will not stand any critical questioning – but which is 'protected' by its place within the effects-religion framework.