Franco Fabbri

Paper delivered at IASPM (UK) conference, 1999. Reproduced online by permission of the author.
For further details, see start of Fabbri’s A Theory of Musical Genres.

It seems very difficult — even impossible — to begin any activity with or about sounds without referring to categories such as ‘kind’, ‘type’, ‘genre’, ‘style’, or to metaphors like ‘field’, ‘area’, ‘space’. Thinking about music, talking about music, making music: all these activities imply reference to a more or less detailed taxonomy, whose structure — not to mention its very existence — is too often taken for granted. Easy access to the world’s diverse musical cultures (and so to even more ‘types of music’) has encouraged expanding and articulating existing taxonomies. However, some of the categories that we use date from more than two thousand years ago, and have been developed differently in different musical and/or national cultures. So, while categories like ‘genre’ or ‘style’ seem to be used mainly to ‘put some order’ and reduce the overall entropy in the musical universe (or, at least, in our talks and writings about music), sometimes they seem to create even more disorder and confusion.

Some questions arise: 1) why do we need musical categories? 2) How are such categories created? 3) Are historical categories like ‘genre’ or ‘style’ useful in all contexts? 4) What is the status of terms like ‘field’, ‘area’, ‘space’, ‘boundary’, and ‘frontier’?

Before we start exploring the usage and function of categories in music, it may be interesting to consider categories in general: that is, the very notion of ‘category’ as a term used in the philosophical and scientific discourse. I am not going to give more than an overview of the various meanings that the term has been assigned to in the history of Western philosophy: it may be enough to point out that the idea of ‘category’ established by Aristotle (and widely accepted for centuries)1 is quite different from the one which was introduced by Immanuel Kant,2 and that the modern usage of the term, mainly by cognitive psychologists, is different again. Today, if one is not a philosopher dealing with some of the historical interpretations of the term, ‘category’ is a class, a set of objects and events, grouped according to some criteria. Much of the discussion amongst researchers (cognitivists, semioticists, philosophers) appears to be about the criteria, about how sets are created: many seem to agree (and if they don’t, they accept the common sense definition) that categories are classes of objects and events, that humans create to reduce the complexity of the empirical world. According to this view, we are able to think just as (while, or because) we are able to create categories: otherwise, we would be lost in the details of infinite multiplicity. As Umberto Eco points out (Eco 1997, p. 123) this is what in ancient philosophy was called — rather than categorizing — ‘conceptualizing’, that is “the problem about how our language (and our cognitive apparatus with it) leads us to speak and think by generalia , or, that we group individuals into sets”.

Eco’s remark is placed in the context of his investigation about how we recognize objects and events: like a platypus, a horse, a mouse, the skyline of a city, a bachelor, Bach’s Second Suitefor cello in a transcription for recorder: a process that Aristotle’s theory of categories appears to give for granted, and Kant doesn’t seem to be very interested in (Eco 1997, p. 124).

Eco suggests that semiosis be involved in this process, and locates the concept of cognitive typeat a node in the process between perception and recognition. A cognitive type is a private set of instructions which allows us to recognize a specific perceptual experience as an occurrence of a particular type : Eco refuses to investigate inside the black box of this process, focusing on the proofs that recognition actually takes place (like when one asks: “bring me the red pencil on the table, please”, and the result is that the red pencil is brought, and not the black ashtray). However, the instructions and descriptions which constitute the cognitive type (in a way that is invisible to anybody), can be shared publicly as a set of interpretants that form what Eco calls the nuclear content, described as “the way we try intersubjectively to make clear which traits make up a cognitive type” (Eco 1997, p. 116). A nuclear content (where content is used in Hjelmslevian terms, as something which is correlated to an expression) is a cultural object, that can be seen as another proof of the existence of a cognitive type — which in turn can be postulated as an individual disposition to produce a socialized nuclear content. Eco also introduces the concept of molar content, as ‘enlarged knowledge’ that includes notions which are not necessary for recognition (ibid., p. 119).

Eco maintains that what modern cognitivists call categories (and that Aristotle would have named predicables) are rather what in natural sciences are called taxa, nested from species to genus (or from orders to classes, or from classes to reigns). Cognitivists’ basic categories are — according to Eco — cognitive types, while other categories belong to “a phase of a more complex cultural elaboration” (Eco 1977, p. 126), and “do not contribute immediately to tell us what a thing is, but rather how it is hierarchically put into order in a system of basic, superordinate and subordinate concepts” (ibid.).

An important point is made by Eco when he examines the status of cultural cases(like ‘cousin’, ‘president’, bishop’, the square root, or events, actions, relations like ‘contract’, or ‘friendship’) as opposed to empirical cases(natural, like ‘mouse’, ‘cat’, ‘tree’, or artificial, like ‘chair, ‘boat’, house’). What all cultural cases have in common is that, “to be recognized as such, they need reference to a framework of cultural norms” (Eco 1997, p. 139). The existence of cognitive types must be acknowledged for cultural cases as well, says Eco; he also suggests that with cultural cases recognition proceeds from the nuclear content (that is, the publicly shared procedures which allow to recognize the object or event) to the cognitive type, rather than the opposite way (like with empirical cases), though he warns immediately that things may not be so simple (ibid., p. 150). Anyway, he comments further on that, though a cognitive type is private, when it is interpreted as nuclear content it becomes public, and a public nuclear content may provide instructions for the creation of cognitive types. So, cognitive types are continuously subject to public control, and the Community (in Eco’s terms) teaches us, step by step, to adapt our cognitive types to other people’s cognitive types. Education is a major agent in this process (ibid., p. 190), and another important factor is art: artists enrich our ability to perceive our environment, by suggesting alternative schemes for perception (ibid., pp. 191-192).

Sometimes Eco comes very close to a discussion of genres, specially when he makes a distinction between generictypes and individualtypes: however, he seems more interested to demonstrate that cognitive types are involved in the recognition of individuals (like George or Robert or Louis XIV) and of formal individuals(like a novel or a piece of music) as well: so his only explicit reference to music in this study is about the process by which someone can recognize a specific aural experience as an occurrence of the type ‘Second suite for celloby Johann Sebastian Bach in a transcription for alto recorder as performed by Frans Brüggen’ (ibid., pp. 183-188). For our purpose, which is to examine much less specific recognition processes and generic rather than individual aural experiences and music events, another part of Eco’s study is more relevant: it’s where he points out the difference between dictionary-like competenceand encyclopaedic competence. The former is based on the ability just to place an entity in a well-defined node of a directory tree, or — according to a more traditional view — in a class of a scientific taxonomy, while the latter implies the knowledge of both the directory structure and the content of the individual directories and files, or, as it is more commonly said, of everything the Community recorded about the entity. The important point is that while linguists maintain that common linguistic competence is purely taxonomic like in a dictionary, Eco suggests that the model for linguistic competence is encyclopaedic (Eco 1997, p. 193-200);3 he also comments that since scientific taxonomies were established only in the eighteenth century, if linguistic competence should be based on a dictionary-like model, then from the appearance of Homo Sapiens until the end of the seventeenth century nobody would have been able to use his own language properly, and Aristotle and Plato, or Descartes and Pascal, would have been talking without understanding each other. Categories that matter both for our ability to recognize things and for our linguistic competence are not like scientific taxonomies: Eco calls them savage categories, in that they are organized and restructured continually at the nuclear content level, grouping objects according to their utility, or to their relevance to our survival, or following formal analogies, and so on (ibid. p. 201). The savage category of ‘insect’ includes spiders, which, according to scientific taxonomies are not insects but arthropods (a class that includes arachnids andinsects).

So, if we accept the term category to mean a generic set of objects or events grouped according to some criteria (which, it must be pointed out again, is not the meaning Aristotle or Kant would give to this word), then we must also admit that we deal with different concepts: a) scientific taxa, which allow — according to our example — to have spiders only in the class of arachnids, and not of insects; b) savage categories, which allow spiders to be insects (as possibly fastidious little animals whose bodies and legs show distinctive junctions); c) encyclopaedic entries, which (if they are really complete and include everything the Community learnt about the object) assign spiders to the correct scientific category of arachnids, while mentioning the folk usage of calling them insects (and, by just some folks and not others, of eating them).

Another rich source of warnings about categories is Lakoff 1987. Most of them are commented in Eco 1997; Eco’s approach, as a philosopher and semioticist (not just as a European, if we consider his continuous reference to the work of Charles S. Peirce) is less that of a dismissal of ‘old’ theories about categories than in Lakoff or in the studies of other cognitivists. Many of Eco’s remarks about some of the cognitivists’ discoveries are acuminate: for example, when he comments the ‘strange’ categorizations of some Australian aboriginals, the Dyirbal, who seem to divide objects in the universe in four classes, which correspond to one of four distinct classifiers (Bayi , Balan, Balam, Bala) that must be used before any term included in one of the classes (Lakoff 1987, pp. 92-93). The Balan class, by the way, includes women, fire, dangerous things, as well as platypuses, echidnas, some snakes, some fishes, most birds, etc. (which explains both Lakoff’s and — partly — Eco’s titles for their books). Eco suggests to compare the Dyirbal’s strange categories with those of two equally strange populations, living respectively in Southern and Northern Europe: the former prepends to any name one of the two words IL (with the variant LO) and LA, which apply to the following categories: IL, for men, kangaroos, bats, many snakes, many fishes, many insects, the sun, a storm, a rainbow, gold, ear, etc.; LA for woman, tiger, some snakes, some fishes, water, the moon, a star, etc.; the latter uses three different words, DER, DIE, DAS, for categories that include the moon (and men) for DER, the sun (and women, and cats) for DIE, and gold and ear for DAS. Eco’s warning is clear: there is a risk to understand grammatical automatic behaviours as categorization processes (Eco 1997, p. 172-173).

Anyway, some of Lakoff’s suggestions are quite stimulating, and probably deserve a more thorough comment than the one allowed in Eco’s perspective. One example is basic-level categorization, as described by Roger Brown and Brent Berlin (Lakoff 1987, pp. 31-38; Brown 1965; Berlin, Breedlove, and Raven 1974). According to Brown, categorization, for a child, begins “at the level of distinctive action” and then proceeds upward to superordinate categories and downward to subordinate categories. This “first level” of categorization was seen by Brown as having the following converging properties:

Berlin and his students and associates go further, and examining folk classifications of plants and animals in a population living in the Chiapas region of Mexico, suggest that the basic level (or the “folk-generic level”) for such classifications is that of the genus, which seems to be a psychologically basic level in the following respects:

Lakoff comments further, basing on the observations of A.J. Cain’s essay “Logic and Memory in Linnaeus’s System of Taxonomy” (1958), that “the heart of the Linnaean system was the genus, not the species. It is the genus that gives the generalcharacteristics and the species that is defined in terms of differentiating characteristics”. And “the genus, as a scientific level of classification, was set up because it was the most psychologically basic level for the purposes of the study of taxonomic biology by human beings” (Lakoff 1987, pp. 34-35). This is a quite challenging point to make, as it seems open to any kind of falsification (and actually Eco doesn’t hesitate to say that “all experiments showed that our knowledge doesn’t comply with this classification”, Eco 1997, p. 157). Anyway, whatever the theoretical framework for basic-level categorization — including Eco’s cognitive types — it is very interesting for a study of categories in music to see that generaappear to have a central function in the functioning of our mind, in the way we perceive the world.

Basic-level categorization was also studied by Eleanor Rosch, who was the first to provide a general perspective for all the specific cases that had already demonstrated weaknesses in the ‘classical theory’ of categories. Her name is mostly associated with the so-called ‘prototype theory’, though in the most recent phase of her studies prototypes are just seen as evidence that there is something wrong or unknown in the ‘classical theory’, rather than proofs of any particular ‘non-classical’ category structure.

A prototype, also called cognitive reference point, is a subcategory or category member that has a special cognitive status — that of being a ‘best example’ (Lakoff 1987, p. 41). If the ‘classical theory’ states that all category members share the same properties, then any demonstration that in a given category some members are more representative than others can be taken as a strong suggestion that the classical model is wrong. Rosch created an experimental framework for such demonstrations, based on paradigms like the following:

Whatever their nature or structure, categories seem to play a fundamental role in our understanding of the world. It would be very strange if — while we spend most of our awake time (and Freud only knows if and how we categorize while dreaming) observing occurrences of types, processing perceptions through cognitive types, socializing nuclear contents, comparing prototypes to nonrepresentative members of categories or vice versa, locating basic categories at the level of the genus — when we listen to music or think or talk about it nothing of this kind (or type, sort, species, class, category...) happened.

In fact, we do use categories for music. We process our perceptions (not only aural perceptions) confronting them with cognitive types — definitely culturalcognitive types — that allow us to recognize what we are hearing (or seeing, or feeling, or touching) as ‘music’, or as ‘music of a certain kind’, rather than ‘noise’. To this respect, Eco’s model seems more appropriate than the concept of basic-level categories, if they are put into the usual framework of superordinate and subordinate categories. ‘Music’ is a superordinate category with respect to a basic level of ‘musics of this or that kind’ only in musicologists’ and musicians’ ‘scientific’ taxonomies: as a savage category, based on the elaboration of the nuclear content by the Community, it may well be at the same level of its apparently subordinate categories (music types, like genres or styles).

We perceive music events both as occurrences of music types (like when we recognize a certain event as an occurrence of the type ‘classical music’ or — if our cognitive types became more articulated via the collective elaboration of their nuclear content — as ‘baroque music performed according to the correct Aufführung Praxis’) and as occurrences of formal individuals, like Johan Sebastian Bach’s Second suite for cello .

It may well be that basic categories for music are located at the level of the genus; to this respect, I’d like to point out that if we find out that this is the level of the categories that some people call ‘genres’ (and I am amongst them), it is not because of the family resemblance of the terms genus and genre, nor because of their ultimate identity in the history of Western philosophy and musicology: it is because genres (at least according to some views) have the qualities that cognitive psychologists acknowledge for basic-level categories. This may not be true for all genres and for all contexts: genres can also be seen as taxonomies superimposed to musics, usually for some ‘bad’ reason (by ‘academics’ or by the music industry, according to folk theories). But if we examine how we recognize music events, how we react to them, how we conceptualize them and organize a social response to them (including a music performance), it seems that most of our effort is concentrated around a basic set of collectively accepted norms that define types of musics, against which we confront — as occurrences — the musical events we are dealing with.

As for categories in general, there are different musical categories at work at the same time (this is probably the same as saying that there are many different musical competences, which is a much more established concept, see Stefani 1978): most of the problems with genres are probably originated at this level, where it is easy to mix genres as folk categories and genres as specialists’ taxonomies — and there are also different specialist traditions). Prototype effects are common, as in all categories related to art: canons are the prototype theory’s ‘best examples’ (or cognitive reference points), and even when taxonomies are based on detailed norms — as it happens both with genres and with styles — there is often an exemplary work or an artist that embody the norms in the most convincing way.

Despite the above mentioned folk theories, then, musical categories aren’t just ‘labels’ applied to musics for obscure reasons related to the profit of professionals (be they musicologists, journalists, record producers, manufacturers or retailers): they seem to exist both at a private level — as cognitive types — and as socialized nuclear content, that is as socialized sets of instructions to detect occurrences of types. In a discussion of musical categories, then, it is not a matter of creating (or commenting) dictionary-like definitions according to which experts can locate an occurrence at a certain level of a directory tree. Rather, I think it is important to comment how our Community’s encyclopaedia describes musical categories, accounting for existing examples and their contradictions. The following is an attempt for an encyclopaedic entry for ‘genre’ (which includes, in the first paragraph, a dictionary-like definition):

A genre is a kind of music, as it is acknowledged by a community for any reason or purpose or criteria, i.e., a set of musical events whose course is governed by rules (of any kind) accepted by a community.

The English term — which refers to any kind of work of art — is more specific than the French word from which it originates; in Neo-Latin languages, as well as in Latin, the equivalent for genre (Italian ‘genere’, Spanish ‘genero’, etc.) means type, kind, class, manner, race, like the ancient Greek genoV As such, the term has been used in the philosophical debate throughout the history of Western civilization, and was very early applied to art and to music. The first clear definition of genoV was given by Aristotle: “What is predicated according to the essence of many which differ specifically”; hence the function of genoV in any definition, which implies two constituents, the genoV and the specific difference (like in “man is a rational animal”, where “animal” is the genoV, and “rational” the specific difference).

Since Aristotle (Poetics) and his pupil Theofrastos (responsible of the first great natural taxonomy, based on the concept of genoV), the idea propagated that what can be predicated according the essence of many works of art (which differ specifically), that is, what makes them similar, worth to belong to the same genoV, is their function; it followed that the style of each work of art should be proper to its genoV (putting it clearly that genre and style are related, but are not the same concept). This is the basis of the theory of the three styles (elevated, medium, humble) which had so many followers amongst Greek and Roman philosophers, orators, poets, playwrights, artists and, as far as it is reported, musicians. The theory emerged again in the Renaissance, and formed the ground for all subsequent discussions: most aesthetic conflicts since then (in music at least since Padre Martini’s critique of Pergolesi’s Stabat Mater, accused to be written in the opera comicastyle of La serva padrona) have been about the decision whether the distinction amongst genres is based somehow on the human nature, or if genres are concepts defined by convention, hence subject to change.

Semioticists and cultural anthropologists in the last decades of the 20th century have set a framework of theories that allowed a wider view of the problem of genres, still in line with its philosophical origins (a matter of definition, first of all) but capable to account of the differences amongst genres — and their increasing multiplicity — in literature, theatre, dance, cinema, music. Not surprisingly, many of these modern theories focus on conventions established within communities, related to performances or their social usages (even for literature: Frye 1957).

In music, genres emerge as names to define similarities, recurrences that members of a community made pertinent to identify musical events: the process can be explicit, like in the proclamation of an aesthetic manifesto, a law, or a marketing campaign, or it can never be declared (see Lewis 1969). Rules that define a genre can be related to any of the codes involved in a musical event (i.e. also rules of behaviour, etiquettes, proxemic and kinesic codes, economic regulations, etc.), in such a way that knowing ‘what kind of music’ one will be listening to (or playing, or talking about, etc.) will act as a compass, helping to choose the proper codes and tools for the participant. Genres, then, can be seen as short-cuts to speed up communication within a musical community, as well as standardised codes that allow no margin for deviation, i.e. no real communication: the rise and fall of progressive rock in the late sixties-early seventies and the subsequent explosion of punk rock are a good example of a recurring process where codes are gradually elaborated, then become too strict and allow only for very predictable texts, then new codes are opposed, and so on. It must be pointed out (as punk is an excellent example for this too) that rules, codes, are made pertinent by the community: what someone sees as the most significant regularity within a certain genre may not be what the community that constituted that genre in the first place saw as its essence (in Aristotelian terms). A hierarchy of codes always defines the ideology of a genre (Eco 1975, Fabbri 1996).

The term originates from the Latin word stilus, the ancient Roman writing tool, connotating ‘a way of writing’. The concept migrated quite soon from literature to music, with meanings that have reflected changing attitudes to norms, rules, conventions, and to originality and individuality. So, while in the Middle Ages style implied a set of models to be adhered to, as they were imposed by tradition or authority, during the Renaissance the concept became more articulated, covering contexts for stylistic diversification as varied as personal, national, functional, hence related to emotions and affect (see Kircher 1650); later the Romantic idea of style as the representation of subjectivity and originality changed (at the end of the 19th century) into the modern concept of a set of rules, a linguistic code. According to Eco (1975) the code which governs an individual text is an aesthetic idiolect, while corpus idiolects or ‘stream’ (or historical) idiolects exist if the same set of rules is applied by the same author to many works, or by different authors when the same idiolect is accepted within a given cultural community, and/or in a given period of time. These different levels of idiolect (style) definitions form a hierarchy of competences, an embedding of hypercodification rules: in other words, style operates at various levels, from the individual work to the genre or historical period, and the difference is a matter of focus or content, rather than quality — the same concept is at work.

As a codified way of making music, which may (or must) conform to specific social functions, style is related to genre, and is sometimes used as its synonym, more often in languages where style is a more common-sense word and genre is felt to be more technical. However, style implies an emphasis on the musical code, while genre relates to all kinds of codes that are referred to in a musical event, so the two terms clearly cover different semantic fields. In fact, though he common usage of style to indicate ‘a way of’ (living, behaving, dressing etc.) is a possible source of overlapping between style and genre, in this broader sense style has an even stronger aesthetic connotation, most probably an heritage of the Romantic meaning.

Much of the Romantic idea of style as personality is still at work in popular music. As a personal feature, style is supposed to be maintained across forms and genres: for example, Ennio Morricone’s composing style can be recognized as such in his film music as well as in pop songs, and even in his orchestrations for songs in the singer-songwriter genre; the same can be said of many performers, though a quality that is often demanded from sessionmen and backing musicians is exactly the opposite, that is, to be able to perform in a given style, and not to impose their own. On the other hand, parody and stylistic quotation are quite common strategies in popular music, providing composers and performers with straightforward tools to create other ‘selves’ (and/or to hide their own): well known examples range from The Beatles’ Sgt. Pepper’s Lonely Hearts Club Band to Bob Dylan’s Nashville Skylineto a massive part of Frank Zappa’s work.

Additional material to reflect (and worry) about with reference to musical classifications was provided recently by the Internet. Whoever tries to find useful information over the World Wide Web understands immediately that classification and similarity are his best allies, and the panic we are facing when exposed to such amounts of information is a good suggestion to understand not the semioticist’s but every man’s and woman’s anxiety to separate, define, classify. In fact, if you go for a CD on the Internet and get into the world’s best known virtual record shop, CDNOW, you will be invited immediately to ‘Browse Our Music Spaces’, which are: Rock/Pop, Jazz/Blues, World/New Age, Urban/Electronic, Country/Folk, Classical. You can also apply to the Album Advisor, a special database engine that will supply a list of records that you will probably enjoy, if you just indicate one that you like (investigating the rules in this database engine would throw some light on current ideas — or at least the webmaster’s — about music similarities). If you look for an alternative in the largest World Wide Web bookshop, then Amazon.com will invite you to ‘Browse Styles’,4 which are Alternative (including sub-categories Indie, Hardcore, New Wave, Punk...), Blues (Modern Electric, Traditional...), Country (Alternative, Bluegrass, Contemporary, Traditional...), Dance & DJ (Ambient, Electronic, Industrial, Techno-House...), Folk (Contemporary, Traditional...), International (Cajun & Zydeco, Reggae, Salsa, Tejano...), Jazz (Avant-Garde, Bebop, Fusion, Latin, Swing...), New Age (Celtic, Environmental, Meditation...), Pop (Adult Contemporary, Easy Listening, Soft Rock, Vocal...), Rap & Hip-Hop (East Coast, Gangsta, New School, West Coast...), Rock (Classic, Hard Rock, Metal, Progressive...), R&B / Soul (Contemporary, Funk, Motown, Soul...), Soundtracks (Movies, Musicals, Television...), Miscellaneous (Children’s, Christian, Comedy, Spoken Word...). And beware: “Looking for Classical? Coming soon--sign up and get notified when we start selling Classical CDs! ”So if we want to know how to classify the classics we have to wait for that particular commercial event.

But we might be interested just in what the Internet is mostly about, information: so why not using a search engine like Yahoo!? There we can find the following URL: http://www.yahoo.com/Entertainment/Music/Genres,5 which will give you a list of links to interesting genre-specific sites, such as: Alternative, Bluegrass, Blues, Cajun and Zydeco, Celtic, Children’s, Classic Rock, Classical, Country and Western, Disco, Dixieland, Dub, Electronica, Flamenco, Folk, Freestyle, Funk, Gospel, Gothic, Indie, Industrial, Jazz, Lounge, March, Metal, Microtonal, Minimalism, New Age, New Wave, Noise, Novelty, Opera, Polka, Progressive Rock, Punk, Ragtime, Rap and Hip Hop, Reggae, Rock and Pop, Rockabilly, Ska, Soul and R&B, Surf Rock, Tejano, Theme Music, World Fusion.

It is worth to point out again that the listings of “music spaces”, “styles” and “genres” largely overlap: webmasters seem to agree that certain kinds or types of music are broadly acknowledged, though they are probably much less interested in understanding how this knowledge is organized. This is what weare interested in, I believe.

Amazon.com suggestion of “style” is consistent with one of the common terms used in English, and with the importance that without any doubt is given to strictly musical elements when distinctions or similarities are made or found between or amongst musics. However, style doesn’t account for other codes that may be of great importance in establishing the meaning of a musical event in a certain context: though it is a flexible and useful concept that spans from such things as “the classical style” to — say — the stylistic consistency of Richard Thompson across genres (i.e. his idiolect as a singer/songwriter and guitarist, see Moore 1998), it can’t explain simple phenomena like: when you realize what music will be performed from how seats (if any) are placed on the floor, or: why does a conductor at the end of a premiere of a piece of contemporary (classical) music always look for the composer in various places in the stalls, when he knows perfectly where he is.

Rituals, etiquettes, proxemic codes, the division of labour, economic procedures and laws, common assumptions based on the music’s function shared in a community, be it a rural community, an urban subculture, a group of people who have the same religious or political beliefs, or just a few bizarre intellectuals who proclaim an aesthetic manifesto: this is the space — the hyperspace — where genres operate. Musical codes are in this space, but this doesn’t mean that style be a subset of genre, or anywhere lower in a hierarchy of musical categories: or, if it is so, it’s in the sense of a subordinate category compared to a basic-level category, that is at a level where more specific information is articulated.

There’s probably more about this. Though any style is a very complex code, people who study musical styles never despair that they will be able to give the most complete description of them (just think of Charles Rosen’s monumental work about The Classical Style); on the other hand, some regularity in the arrangement of musical elements must be found if one wants to talk about style at all. Behind the very notion of style, then, there is an assumption of whole, complete, of necessary and sufficient. We don’t need allof that to recognize and categorize music. In our cognitive experience, we recognize types of phenomena according to partial descriptions, to truncated knowledge. Curiously, but truthfully, genre appears to be a less specific concept than style: as in Eco’s example about recognizing a town after being carried there blindfolded (Eco 1997, p. 189), we know how to recognize instantly a few genres (even without listening to any music), though we would not be able to give someone else a set of instructions to carry out that recognition. We would rather point at prototypes, at ‘best examples’, as interpretants of our own, private cognitive type. Genres are more about beliefs and practice than about theory.

On the other hand, one can give a very detailed list of the norms involved in the definition of a genre, or of a set of genres, like I did in Fabbri 1982b for the ‘System of Canzone in Italy’. But that is more an encyclopaedic definition, which tries to account both for scientific taxonomies and folk categories (or, in Eco’s terms, which deals with the molar content rather than just the nuclear content). And conversely, it must be said that cognitive types for music styles do exist (styles are actually recognized), though in this case we are not at the level of detailed scientific taxonomies. In the end, the difference between genre and style remains in the wider scope of genre, which accounts for ‘non-musical’ properties; both genre and style can be articulated at different levels, covering just the nuclear content and basing on prototypes, ‘best examples’ and ‘family resemblances’, or expanding over the molar content; genre, at least outside the English speaking scientific community where the term is related to historically established taxonomies, is closer to common musical competence (which is less suspicious about non-musical traits), while style is closer to a more elaborated musical competence.

A meaningful effect of these discrepancies emerged during a survey on music consumption and interests (Ala, Fabbri, Fiori, Ghezzi 1985). Interviewees were presented with a list of 88 genres, with the suggestion to indicate those they liked or were interested in, and (if any) those they despised.6 Problems arose when designing the list for classical genres, as some (it could be said in terms the authors couldn’t be aware of at that time) were more basic-level categories, like ‘opera’ or ‘electronic music’, and others needed reference to subordinate, more detailed taxonomies. So — probably aiming at excessive detail — the ‘Lieder’ category was split into ‘classical Lieder’ and ‘romantic Lieder’. The result was that ‘classical Lieder’ (a category that was intended to cover Lieder of the Classic Age, that is the not so widely known Lieder by Mozart and Beethoven) received many more favourable responses than ‘romantic Lieder’, that is Schubert’s, Schumann’s etc., which is what interviewees most probably meant when they said they liked ‘classical Lieder’. Common musical competence sees ‘classical’ as a predicate for all Western art music, while the notion of a classic age and style is elaborated at the level of the molar content, and pertains to specialists’ taxonomies. This confusion of different category levels and competences is quite common, and is somehow enhanced by the easier access to information provided by electronic media and by global communication processes. Whether we enter a record megastore or a virtual shop over the Internet, we are confronted with lists that include an increasing number of music types (whatever they are called), with no apparent care about taxonomic criteria. ‘World Music’ doesn’t seem to include Neapolitan song (which may be intended to belong to the mainstream of pop music, but no explanation is given: see World Music. The Rough Guide. London: Penguin Books, 1994); however, in record shops all over Europe and North America, in the World Music section, under the label Italy, you will mostly find mainstream Italian pop music.

CDNOW’s idea of “spaces” is stimulating, and I definitely support it if it implies that musics are multidimensional cultural entities, which can be represented mentally as objects in an n-dimensional hyperspace. This is the vision suggested by Umberto Eco’s Q-Model for the semantic space (1976), which most likely originates (as an extension) from the structuralist representation of semantic fields as bordering surfaces (see Ullmann 1962), like in the evergreen discussion of terms indicating trees and woods in various languages. However, if we move from this “space” concept down again to a surface model, which is one of the most common mental representations of any classification, including genres, complexity is reduced and we are immediately facing with misleading concepts like “territory”, “border”, “no-man’s-land”.

Again, we are confronted with problems at different levels: that our cognitive apparatus inclines to use interpretants that derive from our experience of space is clear (as the usage of terms like ‘level’ and ‘inclines’ in this very phrase demonstrates — see Fauconnier 1997); however, when we create a spatial mental representation of a concept, we are using a metaphor, we are moving some predicates from one domain to another, and there is no guarantee that there will be a one-to-one correspondence between properties in the two domains. For example, even if we take the most basic-level definition of a genre or style (in terms of nuclear content, that is of the description of the simplest procedures to identify that entity), we’ll probably find a significant number of different properties, like in the matrix suggested by Tagg 1982 for distinguishing ‘folk’, ‘art’ and ‘popular’ musics. If we try to project a similar scheme onto a surface, we’ll find that the spatial representation is impoverished: it doesn’t account for properties that are common to all categories, not to mention all ‘non-classical’ effects (like prototypical effects, degrees of representation, fuzzy sets, family resemblances etc.). A complex universe becomes a map, with clear-cut borders: objects can be either on one side or on the other; to account for objects that do not fit in the representation other metaphors have to be created, like frontiers or cross-overs. This is an area (yes!) where common competence is subject to the continuous influence of a more detailed (though by no means more scientific) competence, that of music critics. It is definitely more comfortable to represent the universe of musics in terms of maps, territories, fields, etc., and to dismiss exceptions as weird objects that live across the borders, rather than account for the differences and nuances that can be found even at the most basic level of musical categories. A history of music criticism might show that such spatial metaphors are not new; however, I have the impression that the usage of terms based in the lexicons of geography and topography became more and more common in the last two decades; some suggestions from Brian Eno’s album covers with maps, paralleled by discourses about ‘possible musics’, might locate (yes, again!) the origins of this fashion (in its current terms) between the end of the Seventies and the beginning of the following decade.7 Fauconnier (1997, p. 9) observes that “mappings become culturally and lexically entrenched, and (...) they actually define the category structure for the language and structure.8 Rather remarkably, although the vocabulary often makes the mapping transparent, we are typically not conscious of the mapping during use, and in fact are liable to be surprised and amused when it is pointed out to us. In such cases, the mapping, although cognitively active, is opaque”. Fauconnier’s concept of mapping is not literal and is much more articulated: however, the comment I just mentioned is in the context of mappings of ‘time as space’, which doesn’t seem to be very different from the one I am discussing. I’d like to point out the close relation between mappings and categories: this means that if spatial metaphors become a norm in the language, then music categories will be restructured according to these metaphors. This is probably already happening.

1. According to Aristotle, a category is the ultimate and most general predicate that can be attributed to anything. Categories have a logical and ontological function: they allow to define entia exactly, by relating them to their general essence. They are: substance, quality, quantity, relation, place, time, position, condition, action, passion.

2. Kant’s categories do not designate essential modes of reality, but the ways we know it: they are a priori functions of the intellect (pure concepts) that determine trascendental conditions for our experience, or the universal forms that phenomena must take to become objects of knowledge. There are twelve categories, under four general titles; every triad shows a thesis, antithesis, synthesis pattern: quantity (unity, plurality, totality), quality (reality, negation, limitation), relation (substance and inherence, causality and dependence, reciprocal action), modality (possibility and impossibility, existence and non-existence, necessity and contingency).

3. See also Eco 1975.

4. I know this will delight my friend Allan Moore, who once said I’m using the term genre to refer to what musicologists (and other people) would call style (Moore 1993, p. 2). I read Amazon’s usage of the term style, instead, as a symptom that this may sound more ‘academically correct’ for a bookshop. Anyway, the point here is classification per se, not according to what principles or parameters.

5. Well, as an authority in classification, Yahoo! is definitely our time’s Aristotle...

6. Results offered a significant correlation in some cases between likers and despisers, as with disco and punk, punk and heavy metal, ‘ballo liscio’ and all rock-related genres.

7. I remember reproaching my friend Richard Middleton for his usage of terms like ‘area’ and ‘field’ (Fabbri 1982); as it is common in these cases, this is more an indication of how the usage of spatial metaphors by other, much less scrupulous critics was already disturbing me at that time.

8. Italics are mine.

Browsing Music Spaces: Categories And The Musical Mind

Browsing Music Spaces:
Categories And The Musical Mind