David Bordwell: Making Meaning

Very much a work in progress…

David Bordwell is a film critic who’s blog, Observations on film art, written in partnership with his wife Kirstin Thompson, is essential reading for those interested in the study of the formal properties of film. A writer of numerous books on film, he is also editor of Post-Theory: Reconstructing Film Studies (1996) with the philosopher Noël Carroll. Bordwell, Thompson and Carroll are at the forefront of a movement among film theorists to move away from the ‘hermeneutic’ or ‘interpretive’ analysis of film rejecting psychoanalysis and aspects of post-structuralism in favour of a more formal and cognitive approach. Their work has been described as ‘neoformalist’ because, like Bakhtin and Vološinov, their work is influenced by Russian formalism.

Bordwell’s Making Meaning: Inference and Rhetoric in the Interpretation of Cinema (1989) is the most indispensable book on film theory of the last fifty years. Bordwell’s project was to deconstruct the theoretical techniques critics – and viewers – routinely employ in the interpretation of film, and the rhetorical strategies they employ to persuade others their reading is important.

Bordwell distinguishes between ‘comprehension’ – the minimum requirement for understanding a plot, for instance, at the level of literal understanding – and ‘interpretation’ – the ‘uncovering’ of ‘hidden’, ‘nonobvious’ meanings. His approach focuses on the role of the spectator:

To assume that sense is ‘in’ the text is to reify what can only be the result of process. Comprehending and interpreting a literary text, a painting, a play, or a film constitutes an activity in which the perceiver plays a central role. The text is inert until a reader or listener or spectator does something to and with it. Moreover, in any act of perception, the effects are ‘undertetermined’ by the data: what E.H. Gombrich calls ‘the beholder’s share’ consists in selecting the structural field. Understanding is mediated by transformative acts, both ‘botton-up’ – mandatory, automatic psychological processes – and ‘top-down’ – conceptual, strategic ones. The sensory data of the film at hand furnish the materials out of which inferential processes of perception and cognition build meanings. Meanings are not found but made. (p2-3)

– David Bordwell, Making Meaning

Meaning is constructed out of textual cues, not created out of thin air; nor is it the result of isolated individuals creating unique and unprecedented interpretations within a social vacuum: interpretation is a psychological and social activity that takes place within particular institutions according to social conventions. Bordwell describes his book as an exercise in ethnography as well as practical poetics.

What I want to do here is summarize Bordwell’s arguments so that readers can be aware of how these strategies influence their own interpretation and evaluation of films (or books, or TV programmes for that matter) and how institutionalised forms of interpretation become naturalised and internalised within interpretive communities.

‘Levels’ of Meaning

Bordwell distinguishes between for levels of meaning:

1) ‘Referential’ meaning. In order to make any sense of the narrative of a film the spectator constructs a concrete world out of the ‘diegesis‘ – the world presented in the film – and the ‘fabula‘ – or ongoing story – which takes place within it. The diegesis can be either ‘extratextual’ – that is, it refers to a real world existing, or which has existed, outside of the film (for example present day Los Angeles, or the 18th Century Paris) or ‘intratextual’ (a world which exists only within the film, such as  the future world of a science fiction film or the fantasy world of a mythic past). At this level, the audience draws upon their knowledge of film conventions, their fundamental conceptions of causality, space and time, and their knowledge of the real world (recognising, for instance, The Statue of Liberty and why it would be out of place rising up out of the sand of a beach). Understanding at this level is the minimum requirement for comprehension.

2) ‘Explicit’ meaning. The spectator assigns an abstract conceptual meaning, or ‘point’, to the diegesis and fabula they have constructed, and this point may be validated by specific textual cues.  A film might have a particular ‘moral’, for instance, which the protagonist learns as the fabula unfolds. The film could have an overt political message, for instance in the films of Oliver Stone or Michael Moore. At this level the film is deemed to have something to say. Again, this level would fall under the general category of comprehension.

3) ‘Implicit’ meaning. Further ‘up’ the levels of abstraction brings us to the construction of covert or symbolic meanings or ‘themes’. Meaning at this level is taken to be implied or ‘spoken’ indirectly. Implicit meaning is  more likely to be a subject of dispute between critics and spectators, and spectators may draw upon extra-textual evidence – such as interviews or references to previous films by the same writer or director – to support their claims. Generally this level of meaning is consistent with referential and explicit meaning though it could contradict those if the referential and explicit meanings are taken to be ironic (as in Paul Verhoeven‘s Starship Troopers, 1997). Where the former two levels of meaning are constitutive of comprehension, implicit meanings are the beginning of ‘interpretation’ proper.

4)  ‘Symptomatic’ or ‘repressed’ meaning. Referential, explicit and implicit criticism assumes the film ‘knows’ what it is doing and the spectator is uncovering intentional meanings. Symptomatic or ‘repressed’ meanings are those the writer or director might not be consciously aware of and may be the result of the psychological (often taken to be psychoanalytical) obsessions on behalf of the creator, or the result of economic, political or ideological conditions in the wider social world. Symptomatic or repressed meaning may run counter to referential, explicit or implicit meaning but this time without irony: as such they are a site of even greater discursive dispute than the previous forms of interpretation. Symptomatic reading is the ‘highest’ form of interpretation – at least as far as critics and academia are concerned. It’s also the most open to conscious or unconscious abuse.

These levels as not as distinct as they may seem when stated so schematically: there may be dispute between those who see a particular level of meaning as ‘implicit’ or ‘symptomatic’, for instance, and disputants may point to particular textual cues or extra-textual sources to support their arguments. Nor does the spectator necessarily progress through levels of interpretation chronologically: comprehension of the story at a ‘referential’ level may depend upon understanding of character motivation at an ‘implicit’ level. And it does not follow that these levels are necessarily of increasing difficulty either: in the art movie, or the narratively complex blockbusters of Christopher Nolan (Memento, 2000, Inception, 2010), referential meaning – comprehension of what is actually going on – may actually be harder to grasp than ‘explicit’ or ‘implicit’ themes.

Interpretation and the Critic

Interpretation is an inferential and rhetorical practice that becomes standardized over time: the strategies are conventional. Critical institutions and other interpretive communities define the ground and parameters of interpretation. Bordwell deftly sketches out a potted history of film criticism.

For Bordwell,

A critic is a person who can perform particular tasks: conceive the possibility of ascribing implicit or repressed meanings to films, invoke acceptable semantic fields, map them onto texts by using conventional schemata and procedures, and produce a ‘model film’ that embodies the interpretation. Though acquired by each individual, these skills and knowledge structures are institutionally defined and transmitted. And though it is possible to abstract a critical ‘theory’ or ‘method’ from individual ‘readings’, and thus to reify that theory or method as a self-sufficient procedure of discovery or validation, employing such an apparatus will not carry any critic all the way through an interpretation. Decisions about cues, patterns, and mapping must still be made by ‘just going on’ as Wittgenstein puts it, and following the tacit logic of craft tradition.

– David Bordwell, Making Meaning

Interpretation is a distinct practice from theorization though there is an overlap. The critic’s job is to

  • Assume the most pertinent meanings to be either implicit or symptomatic or both.
  • Make salient one or more semantic fields.
  • Map the semantic fields onto the film at several levels by correlating textual units with semantic features
  • Articulate an argument that demonstrates the novelty and validity of the interpretation.

In my Making Doctor Who Mean articlesI argue that these practices are not exclusive to the institutions of criticism: that they have become institutionalised in the less formal organisations of fandom. Fans are people who successfully employ these strategies in such a way that they transform the text into something pleasurable and rewarding for them personally, or in such a way that it gains them credibility or prestige within an interpretive (fan) community: interpretation has a social function. I’ll come back to that later though.

Professional critics employ these strategies to gain prestige or .

Non-fans may create a negative interpretation to gain a pleasurable sense of superiority over fans, or to gain notoriety within their own interpretive community. Sometimes they are just being a dick:

Interpretation is the revenge of the intellect upon art…the compliment that mediocrity pays to genius

– Susan Sontag, Against Interpretation

Semantic fields

A semantic field is ‘a set of relations of meaning between conceptual or linguistic units’ (Bordwell, p. 106). Semantic fields describe the area of meaning bound by a sign. Bordwell gives the examples of city/country, a semantic field related by opposite meanings (antonymy),  and city/state/region/country, a semantic field defined by inclusion.

Institutionally appropriate semantic fields are chosen on the basis of their specificity (does the semantic field plausibly fit with textual cues present in a particular film ?) and totality (do the chosen semantic fields cover the film as a whole?).

Semantic fields may form clusters, or themes.

City/country form a common type of semantic field known as a doublet or a binary opposition. These are by far the most common type of semantic field you’ll find in contemporary criticism. Other doublets include man/woman, subject/object, active/passive, living/dead, power/powerlessness, etc.

Doublets can be arranged as proportional series following the logic that a is to b as c is to d, etc. so that – as in much feminist criticism, for instance – man is to woman as subject is to object and active is to passive or  power is to powerless. This is the logic of analogy. be Columns and matrices of binary oppositions can be constructed from such doublets:

Where there are binary oppositions one term is generally taken to be dominant, so man, for instance, is dominant over woman, white over black, active over passive, etc reflecting the dominant values of a particular society. Structuralist critics see binary oppositions as central to ‘Western’ thought – even while constructing a binary opposition between West/East in the process! Presence/absence is seen as central to Western philosophy as the term presence is dominant over absence because absence is what you get when presence is taken away. Psychoanalytically inclined philosophers take this into loopier territory: male is dominant over female because male is the presence of the phallus while female is its absence. I look further into this silliness in my article on Laura Mulvey’s ”Visual Pleasure and Narrative Cinema”: suffice it to ask here why female is not the presence of the womb and male is not the womb’s absence? I’m not denying that men occupy most positions of power but this is hardly restricted to ‘Western’ society (a comparison with ‘Eastern’ societies would soon disabuse critics of this notion) and psychobabble about ‘absent’ phalluses help nobody. It’s not even necessary for both sets of the binary pair to be present in the film for them to be significant.  John Carpenter‘s The Thing (1982), for instance, has no female characters: nonetheless some critics would find ‘woman’ to be a ‘structuring absence’. This allows them to draw analogies between woman and the Thing itself, licensed, as ever, by dubious Freudian concepts. You don’t need to be Kurt Gödel either to realise that the more comprehensively you break the film down into binary opposites the more inconsistent the proportional series becomes: human may be opposed to machine, for instance – or just as logically to animal. When critics start claiming that Arnold Schwarzenegger is feminised in The Terminator (1984) based on such proportional series – as one critic does in Terminator and Philosophy – you know something has gone drastically wrong!

A more structured form of proportional series is that of the Greimas semiotic square. The Semiotic Square is derived from Aristotelian logic and formed by an initial binary relationship between two contrary signs, A and E, the universal affirmative and the universal negative. A is the assertion/positive element, of the form usually denoted in philosophy as ‘every S is P’, and E is the negation/negative element in the binary pair generally denoted as ‘no S is P’. Aristotle states that to every affirmation there corresponds exactly one negation, and that every affirmation and its negation are opposed such that always one of them must be true, and the other false.

The example Bordwell uses is taken from James Kavanaugh’s analysis of the semantics of humanness in Ridley Scott‘s  Alien (1979), where A represents Ripley (Sigourney Weaver), the assertive/positive term, in binary opposition – or contrary – to the Alien, the negation/negative term, at E. I represents the particular affirmative, which includes some aspects of A (‘some S is P’) while O is the particular negative (‘some S is not P’). In the Greimas Square I is complementary to A, not oppositional. Kavanaugh fills this position with Jones, the cat, which, while not human nevertheless portrays some human characteristics. O is the contradictory position, sharing non-human aspects of the Alien. The Greimas Square is highly ‘logical’ but hardly a complete description of the film – is the cat really a more semantically significant figure than Dallas and the rest of the crew? This is how movies are analysed on the planet Vulcan. (I look no less favourably at John Fiske’s semiotic square of the Doctor Who story The Creature from the Pit in .)

Binary oppositions such as good/evil or sanity/madness collapse easily under scrutiny in any case: nobody is either perfectly good or utterly evil, completely sane or totally mental. There is a graded scale between these polar opposites which will come as no surprise to anyone who doesn’t think in binary opposites in the first place; nevertheless, critics who ‘discover’that films don’t break down into patterns they have tried to impose upon them will attribute all kinds of significance to this ‘discovery’ depending on whether they are sympathetic to the film or not: a seemingly ‘reactionary’ film that the critic just happens to enjoy will be ‘ruptured’ by a repressed ‘progressive’ subtext while an explicitly ‘progressive’ film the critic doesn’t like much will turn out to be harbouring ‘reactionary’ ideology. Jaques Derrida made a ‘philosophy’ out of deconstructing binary opposites.’

Schemata and Heuristics

‘Schemata’ are knowledge structures. Schemata can be ‘bottom up’  (or ‘automatic’) or ‘top down’. Schemata can be primed. ‘Heuristics’ are reasoning routines which connect textual cues, schemata and semantic fields. They may be culturally widespread or domain-specific to the critical institution. Making inferences about character’s state of mind, for instance, is a widespread human characteristic; the assumption that the film is ‘about’ these states of might be a heuristic specific to the interpretive community.

Bordwell identifiesfour main schemata, the ‘category’ schema, the ‘person’ schema and two textual schemata.

The category schema places the film within a particular category, such as fiction or non-fiction.

One of the most common category schemata is that of ‘genre’. Definitions of particular genre are notoriously difficult to pin down – I have a crack at defining science fiction elsewhere – but ‘genre’ is so routinely deployed as a schema that it is rarely questioned. Certain films stand as exemplars or prototypes of particular genres (Stagecoach or High Noon for the Western, Dracula or The Exorcist for supernatural horror, Forbidden Planet or 2001: A Space Odyssey (1968) for science fiction, The Thief of Bagdad or The Lord of the Rings for fantasy, The Maltese Falcon or Chinatown for film noir) but there are also genre ‘hybrids’ such as the science fiction Western (Westworld), science fiction horror (Alien, 1979), The Fly), science fantasy (Star Wars), the supernatural Western (High Plains DrifterPale Rider)  or the science fiction film noir ( Blade Runner, 1982). In addition some directors developed a distinctive style which has been imitated by others such that as it is possible to speak of the ‘Hitchcock‘ film or the ‘Hitchcockian‘ film. Genre is perhaps best understood as a core/periphery schema with ‘ideal’ types clustered towards the core with more difficult to define examples floating around the periphery. Note that genre prototypes are not generally produced as exemplars: film noir did not ‘exist’ as a genre until the Fifties when French critics retrospectively imposed the term on films that had hitherto been classed as crime thrillers, albeit crime films which shared a similar expressionistic style and recurring characters such as the doomed hero and the ‘femme fatale‘.

The ‘person’ schema, or ‘personification’, is the attribution of human or ‘folk-psychological’ characteristics to aspects of the film. Least controversial of these, at least to the popular audience, is the attribution of human characteristics to characters in the film: indeed it would be impossible to comprehend a movie at even the ‘referential’ level without some attribution of motivation, emotional complexity or rational agency to the protagonists. Of course, avant-garde critics regard the rational agent with suspicion (I’ll discuss this in Mystifying Movies) but, as we’ll also see, they aren’t terribly consistent about this).

Characters become the bearers of semantic fields such as joy or sadness, love or hate, freedom or captivity; to the symptomatic critic they may embody sadism or voyeurism. They may also incarnate abstract or symbolic characteristics: ‘what a character is or has can be translated into what the character means’ (Bordwell, p. 154). If characters are in conflict then the semantic fields they embody can also be seen to be in conflict.

The ‘persona’ schema can also be applied to the film maker: generally speaking critics regard the director as the author of the film (on TV it would more likely be the writer). Personification of the film maker was particularly common among auteur critics but also appears in the writing of the avante-garde when explicating modernist films where the film maker is supposedly laying bare the process of film making: the director is granted the status of ‘rational agent’ otherwise denied popular film makers, their characters or the popular audience.

Some directors make cameo appearances in their own films (Hitchcock in most of his movies, for instance, or Francis Ford Coppola in Apocalypse Now) but elsewhere a character or the narrative or a stylistic treatment ’embodies’ the film maker’s intentions or repressed urges. Interviews or (and this wasn’t the case when Bordwell’s book was written) directors’ commentaries give further insight into the director’s mind.

Narrative and cinematic style may be personified independently of the film maker: the term ‘narrative’ implies a person providing narration. A film may be described as  ‘witty’ or ‘cynical’ or ‘paranoid’ – human characteristics it clearly cannot have. Critics increasingly focussed on ‘the camera’, an abstract construct unrelated to the simple mechanical device which captures light on film: this anthropomorphised construct is said to ‘select’ what it ‘presents’ to the audience, to be ‘obsessed’ with particular details, or to be the bearer of the ‘sadistic’ or ‘voyeuristic’ gaze. These qualities can then be mapped back on to the film maker or the audience.

And finally the ‘person’ schema can be projected on to the audience itself. The most pervasive heuristic here is that of ‘identification’. ‘Identification’ is a link schema connecting the audience with, typically, a character (what Christian Metz calls ‘secondary identification’) or with the camera’s ‘point of view’ (‘primary identification’). This, according to avante-garde film makers, leads to political quiescence; following the lead of Bertolt Brecht, the avante-garde attempts to undermine ‘identification’ by producing ‘distancing’ or ‘alienating’ effects. Needless to say, ‘identification’ is yet another concept borrowed from psychoanalysis without any empirical justification.

There are two textual schemata.

The critic ‘maps’ semantic fields onto textual cues with the aid of institutionally permissible heuristics  and  schemata.

Bordwell’s cognitive approach, with its emphasis upon human perception, cultural experience, schemata and heuristic procedures, institutional norms and practices, leads us to the rather scandalous conclusion that ‘meaning’ is not intrinsic to the film but extrinsic, and constructed by the spectator. This is either disturbing – particularly for those who see film primarily as vectors for dominant ideology and the like – or profoundly liberating. My money is on the latter.


Critics also have to persuade others their interpretation is worthy of discussion and to this end they apply rhetorical moves largely unchanged since the time of Aristotle (384-322 BC) and Cicero (106-43 BC). Yep, its that old.

The critics first job is to frame their interpretation to meet the readers expectations – even if they are then to challenge their preconceptions.

”The speaker must frame his proofs and arguments with the help of common knowledge and accepted oppinions.”

– Aristotle, The Rhetoric

The speaker must frame

The two main forms of rhetoric are inventio, dispositio and elocutio.

Inventio refers to

Disposito is Latin for organisation or arrangement. An explicatory critic will generally organise their argument around an intuitive experience of the film while inferential or symptomatic critics will mix abstract theories drawn from authorities like Freud, Lacan or Althusser with claims the film demonstrates the truth of those theories.

A typical interpretation follows Aristotle’s schema, as elaborated by Cicero:


  • Entrance: An introduction to the issue.
  • Narration: The background circumstances (a brief account of the issue)
  • Proposition: A statement of the thesis to be proven.


  • Division: A breakdown of the points that support the thesis.
  • Confirmation: The arguments under each point.
  • Confutation: The demolition of opposing arguments

Conclusion: A review and emotional exhortation.

Bordwell notes that these components may be reordered – confutation, for instance, may appear near the beginning, or not at all – or scattered throughout the length of the argument.

Elocution are the stylistic elements of argument: correctness, clearness, appropriateness, and ornament.


Bordwell’s cognitive approach has been met with a somewhat hysterical backlash among those who’s theories remain rooted in psychoanalysis and Althusseran Marxism, including Colin McCabe and Slavoj Žižek.


  • Bordwell, David (1989) Making Meaning: Inference and Rhetoric in the Interpretation of Cinema
  • Bordwell, David & Noël Carroll (1996) Post-Theory: Reconstructing Film Studies
  • Fiske, Jonn (1984) ”Popularity and Ideology: A Structuralist Reading of Dr Who” in Rowland, Willard D Jr and Watkins, Bruce (eds) (1984)
  • Kavanaugh, James H. (1980) ”Son of a Bitch: Feminism, Humanism, and Science in Alien”, October 13
  • Sontag, Susan (1966) Against Interpretation
  • Rowland, Willard D. Jr. and Watkins, Bruce (eds) (1984) Interpreting Television: Current Research Perspectives
  1. Eric Prince says:

    Just want to say that when you talk about Kavannah semiotic square the term that is in opposition to Human (Ripley) is anti-human and not the term non-human. Therefore the cat is at the right place. Also the question with this took is not to determine the place of the element within a narrative but to show some semantic fiels. You can even identify some elements that are not (at least explicitely) in a film but would figure in a complete semantic field.

  2. […]  David Bordwell codified this approach nicely in his work on film analysis (here’s a useful blog post summarizing […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s