The Cognitive and Philosophical Approaches to Horror Films

Aaron Smuts
Aaron Smuts

Philosophical work on horror has been predominantly focused on the horror film, though little of what has been written on horror is medium specific. It is just that the overwhelming majority of examples in the literature are movies (Schneider and Shaw, 2003).

Here, I continue the trend. This entry concerns a relatively small topic in a sub-area of film theory that is often called the analytic-cognitivist tradition. This tradition has no clear unifying, positive doctrine (contra Sinnerbrink, 2011: 4–5). Instead, the tradition is best described negatively.

Two aversions are important: first, film theorists of the analytic-cognitive stripe exhibit a pronounced suspicion of psychoanalytic accounts of mental activity, preferring instead explanations from contemporary cognitive psychology.

Second, analytic-cognitivist film theorists tend to have a strong antipathy to much of what is called continental philosophy, whatever that might be. They align with the dominant approach in the English-speaking-world, that of analytic philosophy (Carroll, 1996).

Although these labels are much maligned, there are important differences between analytic and continental philosophical practice. This is not amere sociological divide (Leiter, 2013).

Analytic philosophy is primarily problem driven, rather than book or figure focused. One does not do philosophy in the analytic style through another philosopher. No, one simply does philosophy.

Similarly, analytic philosophy emphasizes work on live problems, not on what others have had to say about the issues. Analytic philosophers tend to address small problems rather than aim for systematicity, though there are lots of exceptions.

The analytic tradition emphasizes clarity of thought and rigorous argumentation, not textual interpretation. Stylistically, the traditions differ greatly, though there is a lot of bad writing on both sides of the divide. These are crude, but characteristic differences between the two schools.

At its worst, analytic philosophy is logic-chopping legalistic philosophy of no clear relevance to larger human concerns. But work in the tradition need not be so uninspiring. The caricature is undermined by some of the work we find on horror.

Four main issues have occupied centre stage in the analytic-cognitivist work on horror: What is horror? What is the appeal of horror? How does it frighten audiences? Is it irrational to be scared of horror fiction?

The first question asks for a definition. This is clearly the driest question in the bunch. But it has important implications for how we answer the others, particularly the second question. The appeal of supernatural horror and slasher movies might be very different. There might be no common source of appeal between the two. If so, only a definition that excludes slasher movies would allow for a general account of horrific appeal. We will look at just such a definition in the next section.

The second question is closely related to the general problem in the philosophy of art, a problem that is sometimes called the paradox of tragedy or, as I prefer, the paradox of painful art. Why do people seek out horror films, melodramas, sad songs, bleak conspiratorial thrillers, and works in other genres that arouse unpleasant emotions? It is puzzling that people so readily subject themselves to horror films that arouse fear and disgust. How can we account for this?

The first question also has important implications for the third and fourth. These both concern, what has become known as the paradox of fiction. We can state it as a question for now: If audiences know that no one is really threatened by the monster in ‘Jeepers Creepers’ (2001), can they, nevertheless, feel genuine, rational fear? Surely, it must be irrational to fear a monster that one knows is merely fictional! The rest of this chapter is structured around these four questions.

Years ago, before Netflix and Amazon streaming, there were these curious places called video stores. They were for-profit lending libraries holding movies. Most video stores were arranged in a predictable manner. Numerous copies of new releases were grouped in one area, but most of the rest of the store was arranged by genre: comedy, drama, action, horror. When we ask, “What is horror?” we are trying to determine the features a work should have in order to be properly classified under the horror section. We might say that we are looking for the essence of horror. We want a definition of horror.

If you asked a video store clerk, “What is a horror movie?” you would have likely received a snide answer: “Look over there. See that movie: ‘Suspiria’ (1977) is a horror movie.” The irritated clerk would have offered you an ostensive definition — a definition by pointing. But that is not what we are looking for.

We are not looking for instances of horror; rather, we are wondering what makes it appropriate to classify all those instances under the same category.We want a real definition. A real definition tells us the properties that an object must have in order to be a work of horror. More precisely, a real definition lists the necessary and jointly sufficient conditions an object must have in order for the concept to apply. It tells us both what properties are required and what properties in combination are enough.

In the philosophy of horror, we find very few definitions of horror. The most well-developed definition is one proposed by Noël Carroll (1990). According to Carroll, a work of horror must include or at least suggest the presence of, a monster. This is key. By “monster” he has something very specific in mind: a monster is a fearsome creature whose existence is not acknowledged by current science (presumably in the world of the fiction). A silly, harmless ghost is not a monster on this conception. It is not a monster because it is not fearsome (Carroll, 1999). Typically, monsters are also categorically interstitial. They do not fit into our conceptual scheme cleanly, occupying something between categories — human and wolf, or animal and vegetable as in ‘The Thing’ from ‘Another World’ (1951).

Horror films not only include monsters, they are designed to arouse fear and disgust directed at a monster. The fearsomeness of the monster arouses fear. The categorical interstitiality arouses disgust. This combination of reactions directed at a monster is what distinguishes horror from all other genres. Thrillers evoke fear and sometimes disgust, but not at a monster. Gross-out comedy might arouse disgust and sometimes fear, but, again, not at a monster.

Science fiction may include creatures that are not acknowledged to exist by current science, but they are not monsters in the fiction. In the world of ‘Star Wars’ (Lucas, 1977), Chewbacca is acknowledged to exist by the leading scientific authorities (though not in our world). And he is not fearsome. Hence, he is not a monster. No, he is just the furry pal of Han Solo. That is why ‘Star Wars’ is not a horror movie. But ‘Alien’ (Scott, 1979) is. The face-huggers are fearsome! Their mother, more so. Monsters make horror.

Carroll’s definition properly excludes comedies, thrillers, fantasy, and science fiction, and it accurately includes a wide array of what we classify as horror fiction. Any number of classic examples fit the theory perfectly. For Carroll, it is fair to say that the paradigm of a horror movie is something akin to ‘Bride of Frankenstein’ (1935). This is indeed a good model. But many worry that his definition is under-inclusive.

As Carroll notes, his definition excludes movies such as ‘Psycho’ (1960), ‘Silence of the Lambs’ (1991), ‘Halloween’ (1978), and maybe ‘Jaws’ (1975). Norman Bates is not something unaccounted for by contemporary science. Nor are Hannibal Lecter, Buffalo Bill, and Michael Myers. They are all species of psychopaths. Our prisons are full of them.

In reply to this worry, Carroll argues that his theory can account for why we are tempted to classify these kinds of films as works of horror. They arouse fear and disgust directed at something that is monster-like. Norman Bates is “nor man, nor woman.” He is categorically interstitial and disgusting when dressed as a mother or carrying around her skeleton.

So, Carroll concludes, his definition is not uninformative. It tells us why these are edge cases. But these edge cases are just that, edge cases. They do not technically fall under the category when we draw a clear border. So be it.

But few have found this reply palatable. The titles above are probably some of the first that we would list if asked to name some key horror movies. Carroll’s definition appears to exclude too many central cases.

The ‘Silence of the Lambs’, or better, ‘Manhunter’ (1986) is not an edge case. It is a paradigm of the genre. But it does not include a supernatural monster. It does not even suggest one. The problem is that a successful definition must account for the paradigm cases. Carroll’s does not.

If we accept the force of this objection, there are roughly two options available: reject the requirement that horror films must feature a monster or reject the supernatural conception of what it is to be a monster. Since it is hard to imagine a horror movie without at least a suggested monster, the second option seems more promising.

There do indeed appear to be more kinds of monsters than just the supernatural type. Slasher movies, such as ‘Halloween’, feature naturalistic, or realistic, monsters. Michael Myers, for instance, is not supernatural. He is mean, creepy, and very hard to kill, but not supernatural. Nor are the hillbillies in ‘The Texas Chain Saw Massacre’ (1974). It looked at this way, horror appears to be a genre with two main subtypes, supernatural horror and realist horror.

Unfortunately, I know of no plausible way to precisely characterize realist horror monsters. That is not to say it cannot be done, but I do not know how to do it. They are evil, but this is not sufficient.

The presence of the evil Darth Vader does not make ‘Star Wars’ a horror movie. Nor does Amon Goeth (Ralph Fiennes’ evil Nazi character) make ‘Schindler’s List’ (1993) a horror movie. Evil is not sufficient, nor does it seem necessary. The monsters in ‘Alien’ are not evil. The chief monster is simply a mother defending her offspring. They are parasitic, ugly, and very dangerous. But I am reluctant to call them evil. Are birds of prey evil or simply dangerous to little lambs?

There are few competing accounts of what it is to be a monster in the literature. They are all failures. For instance, consider the idea that a monster is something that has failed to achieve its natural end, that has in some way subverted its nature (Yanal, 2003). This will not do. The monster in ‘Jeepers Creepers’ is doing exactly what a creature of its kind does. It realizes its nature effectively. The same holds for the face-huggers in ‘Alien’.

Since the prospect of a formal definition rides on our ability to define what it is to be a monster, we are left at this point without a workable real definition. We are not in a much better position than the irritated video store clerk who flippantly offered an ostensive definition. But we have come to an important insight: the contemporary horror genre, as popularly understood, has both supernatural and realist traditions. This will complicate things as we proceed.

Share on facebook
Share on linkedin
Share on twitter
Share on reddit
Share on pinterest
Share on whatsapp
Notify of
Inline Discussions
View all discussions
Limited Time Promotion

Up To


Stay up to date in history, music, literature, cultural and other studies such as theatre and performance, health, lifestyle, philosophy and religion and get early notice of new content, invitations to exclusive events and special offers and promotions.