1/3/2002

Final Exam

1. ?What music expresses for me is not thoughts too indefinite to clothe in words, but too definite.?
(a) Mendelssohn?s remark implies first and foremost that he considers music a different kind of entity than any language consisting of words. This viewpoint aligns with the theories of music proposed for instance by Langer, in which music shares many of language?s symbolic properties but lacks some essential characteristics we require for a language proper. However, for Langer, the missing elements are the fixed connotations attached to words; Mendelssohn, meanwhile, here argues that music has more potential to express definite thoughts.

(b) Without some context to determine what kind of thoughts Mendelssohn had in mind, it is difficult to evaluate his claim in the general case. On the other hand, interpreting the excerpt as an existence claim for applicable thoughts, merely one example would give his viewpoint validity. In fact, one such kind of thought comes to mind readily: emotion.

Let us first clarify the concept of emotions under discussion and separate them from the broader categories of feelings and thoughts. With respect to one?s preferred theory of mind, we can say that thoughts comprise all the various states of mind that are internally intelligible as concepts, predicates, feelings, and so forth; these can be as complete or fragmentary and as abstract or concrete as we can imagine. Feelings, meanwhile, are thoughts that affect our judgments: we thus exclude both the concept ?Hamlet?s sadness? and the predicate ?sad? itself, but retain ?Hamlet was sad? for its impact on our evaluation of the play. They are in this sense necessarily more definite than general thoughts, because otherwise they could not influence our rational processes.

Although my previous classification may seem a bit arbitrary, the distinction I feel essential between feelings and the subset we call emotions is much more intuitive: emotions concern our inner selves. While reading Shakespeare may induce various emotions, no statement about his stories is itself experienced the way we do an emotive thought. Even ?I am sad? fails to be more than a feeling ? despite being complete, (relatively) concrete, and personally relevant ? because it does not directly interact with our conscious personality the way most people would describe their emotional reactions. Aristotle called emotions those feelings that are attended by pain and/or pleasure. Though I feel this particular definition somewhat circular, it?s clear that emotions must be tied to something more primitive than purely cognitive feelings(1).

So far I haven?t drawn any conclusions I feel would be rejected by the average person, or even the average philosopher. Nonetheless they are sufficient to give weight to the first part of Mendelssohn?s claim: that there are thoughts more aptly expressed by music than is possible with language. We?ve already seen the inadequacy of language in this context, since I was only able to give counterexamples; I could have listed some true emotions but these would simply be identifiers, not descriptions ? much less instances of the visceral impact necessary for a thought to be an emotion. While the orthographic arts can achieve expression on the descriptive level, even a very arrogant writer or poet would concede that their work requires an attentive reader for it to actually generate emotion. Meanwhile, music?s expressive power in this regard is such that it can captivate even a passive listener; although few moviegoers listen to the score with a critical ear, its emotional impact can be dramatic.

Thus, this property of music reflects a fundamental compatibility with emotion. That language attempts to identify, then describe, then instantiate an emotion reveals only the relative strengths of language, not anything about the nature of emotion. Most people probably feel their emotional reactions take the opposite course ? i.e., that they are experienced first, and only subject to description and identification when linguistic articulation demands. Indeed, the use of language to express emotions in this way is usually considered limiting in an essential way, failing to capture the nature of the emotions. Music, conversely, fails in the task of pure identification (Langer?s observation), but can ? as illustrated ? convey emotion unlike any written work. I won?t offer here a theory of why music has this potential, but granting that assumption I?ll note that this argument aligns with not only popular notions but with the definitions given earlier(2).

However, to prove the second part of Mendelssohn?s claim ? that emotions are in some respect more definite than the thoughts readily expressed by language ? we must abandon some ?common sense? assumptions. We?re tempted to treat emotions as more ephemeral, more indefinite in Mendelssohn?s terminology. I?ll grant that they?re more transitory than most other thoughts, but my previous logic would support just the opposite claim of definite-ness. When distinguishing feelings from thoughts, I noted the ability of feelings to affect judgments, and reasoned that they must be definite in order to influence the decision-making process. Specifically, for any thought to be given rational consideration, one could postulate it must be fixed and explicit.

How then would emotions fit? They seem to be the most stimulating of any feelings; these effects should allow us to conclude that they are more definite despite their admitted evanescence. Note that we call judgments ?irrational? when based on factors considered unnecessarily short-lived ? the class of feelings known as opinions, for instance, seems a poor basis for ?objective? decision-making. Decisions rendered ?irrational? by non-fixed thoughts violate pure logic to be sure, but we can safely conclude they are still governed by the same process. An overly vague or fragmentary thought, on the other hand, renders logical analysis impotent instead of merely producing ?wrong? results. Thus, it?s the remaining factor of explicitness that governs definiteness in the sense we?ve been exploring.

What remains is to show how emotions can in fact be explicit. Although counterintuitive at first, we?ve already seen one (misleading) reason for the common perception to the contrary: emotions are not adequately expressible in language. Clearly, though, this is no reason to dismiss them as necessarily vague. If anything, the sense that putting emotions into words somehow reduces them ? in a way other forms of expression like music do not ? should indicate that here language is the cumbersome, inexplicit object, not that which we?re attempting to describe. Meanwhile, the fact that music, itself a very complex yet orderly phenomenon, seems able to express emotion suggests that emotion is not fragmentary, discontinuous, or otherwise disruptive to the thought processes described earlier ? in short, that it is definite.

To what extent to I agree with all of this? The thoughts > feelings > emotions framework I constructed seems logical enough, if perhaps unwarranted. I believe I?m justified in the various ways I narrowed the target property(3). The assertion that emotions meet even the limited target of explicitness is a bit more tenuous; the strongest part of that argument is the rationalization for our usual belief, which is of course not really evidence for the desired claim. Accepting for a minute that music does have the special relationship to emotions I allege, there are more possible ramifications than the desired one: emotions could be equally complex, but still amorphous; they could be perfectly explicit, but still invalidate the decision-making process in some other way; and so on. Of one thing I am convinced: definite & indefinite, vague & explicit ? all of these are pretty ill-defined terms (which ? fortunately or unfortunately? ? means that if nothing else I can twist them into compliance).

(c) The essential features of music that allow it to function as an expressive vehicle for emotions are present in all art forms; one might even argue that this ability is a prerequisite for a creative activity to be considered ?art.? Some of these features are even common to language: for instance, a previous argument [that music?s ability to depict emotions implies that they are definite] relies on the fact that music itself is not vague. I.e., like language, every component of a musical work is fundamentally exclusionary. The fact that a piece is in 4/4, for instance, means it is not in 3/4 or 2/4 or any other meter, just as the fact that a novel is in English means it is not in Spanish et al.; such well-defined features exist down to the level of the third note of the horn part?s being an Eb excluding any other note, just as the third word of some chapter?s being ?the? excludes all other words. All art forms seem to share this trait without issue: a blue sky in a painting it is not orange, the instruction to raise one?s arms in dance excludes the opposite, and so on.

However, some ways in which music differs from language must also be shared by the other arts, or language itself would qualify! Here I think the difference noted by Langer is critical: language?s inability to express emotion in a way compatible with our perception of it (i.e., with description being subservient to direct experience) stems from its fixed associations. Because its static meanings are inescapable, it can?t convey the dynamic, visceral impact of emotion, although it?s more than adequate for identifying and categorizing the related thoughts. No other art suffers this defect as completely, although strictly representational paintings and sculptures will share some of the difficulty in wrenching the casual observer away from the obvious connotations to their emotional underpinnings.


2. I?m tempted at first to analyze expressive properties without regard to a musical work itself; in the end I return to this view. The most obvious ways to characterize the chief differences between our old categories of primary properties (?the statue is 6 lb.?), secondary properties (?the statue is smooth?), and tertiary properties (?the statue is sad?) are externally verifiable, for instance the probability my classmates would all agree on a description(4). The degree to which the properties of a work remain unchanged between performances would distinguish these categories in a similar way; thus, I would be arguing that a work?s properties are isomorphic to external factors. Expressive properties clearly belonging to the third category, they would have (speaking trivially) the most potential for change, but more importantly be dependent on the differing perceptions of each performance (regardless of to what degree the perceptions corresponded to actual differences in the way the work was performed, I might add).

However, we want to somehow be consistent in also saying that at least some of a work?s properties are inherent, thus independent not only of particular performances but of any performance at all. If I wrote a piece in F minor, it would still be in F minor even if it never premiered. As a possible solution I propose a thought-experiment: imagine a machine (perhaps a piece of software) that could predict with perfect accuracy the percentage of listeners who would find the work sad, were it performed. Given only the score, can we now be as ?objective? in a sense about the piece?s sadness as we were about its key? Even aside from the fact that we?re totally ignoring the variations between performances, I have serious reservations. For one, it only highlights if anything the fact that sadness is still a tertiary property according to the old benchmark, since the percentage of the audience that will identify with subjective predicates like sadness will never be as high as the near-certainty with which people can agree on formal properties. Furthermore, it dodges the underlying philosophical issue: does a claim about sadness have meaning at all unless the emotion is actually evoked? Even if we could move from predicting percentages to specifics (e.g. that John & Mary would find it sad but James would not), this process would still be quite removed from a performance.

The first objection is defined away easily enough: if we say that tertiary properties must necessarily exist on a continuum with every piece being neither completely sad nor devoid of sadness (vs. the way we say definitively whether a piece is a violin solo or not), we haven?t really lost any descriptive power. To evaluate the second, let us at the outset be clear as to the distinction between perceiving an emotion and experiencing it, one drawn by theorists since at least Mattheson. I also break rank with the essays I?ve read so far and assign the former?s study to theories of representation in music(5). After all, the domain of artistic representation is thought, and there seems little trouble in calling ?thoughts? the abstract feelings sensed in music yet not undergone(6). Thus we concern ourselves now with the implications of our thought-experiment on the way music moves us.

From this viewpoint I find that my second objection holds: the existence of a ?black box? that predicts emotional reactions would not imply that these expressive properties are at all verifiable in the work, because they only make sense in the context of a performance. Moreover, this incompatibility stems not from some deficiency in our thought-experiment, but is merely highlighted by it; we shall see that the nature of a musical work itself proves inadequate to contain the emotions it stirs. I?ve already rejected the pure ?emotivist? view, that recognizing sadness inherent in a work equates to experiencing it(7). Meanwhile, looking single-mindedly for an ?Uncle Charlie? is no real alternative for our purposes; it may be simple beauty that moves some people, but this theory sheds no light on whether that beauty is inherent to the work, the performance, or some combination.

Regardless of whether his psychological underpinnings prove true, I think Meyer pointed us in the right direction. By hypothesizing that ?emotion is evoked when a tendency to respond is inhibited,? we notice the importance of time. This Law of Affect shows how emotions can depend on music progressing in sync with our thoughts, the tones only being perceived before or after others, allowing for our mind to change between them, in his case by inviting or suppressing a response. I don?t believe this phenomenon isolated to critical listeners conscious of the expectations raised and realized (or not) in music; I think it necessary for any emotional experience that a specific performance be experienced in time, the pieces of a work heard in sequence, because a stirred emotion is an essentially dynamic thing.

I mean this in a deeper sense than Meyer?s distinction between emotions and moods: an emotion is not just evanescent, but can only experienced (in the sense we?ve limited ourselves to dealing with) as progressing in some sense. As I think I alluded to in #1, emotions are the thoughts that (for better or worse) have the greatest effect on our choices; there is attendant with the strongest form of each a desire to do something ? jump for joy, mourn in sadness, extract revenge. Even our terminology reflects it ? we say we are ?moved,? i.e. put into motion. Accordingly, when the next note is struck, our mental state is no longer as it was before the previous one(8).

Contrast this with our notion of a musical work: it is separated from our concepts of a performance, a score, and so on by its timelessness. Conceptions vary, but in the strongest sense a work can exist in the composer?s mind without ever being written down much less shared. Even with our black-box ideal for extracting as many ?objective? properties as we like from it, we still see a static score being taken as a whole and considered without regard to the linear way an audience would hear it. To use Langer?s term, the ?creation of virtual time? can never realized by a work alone. In this way they differ fundamentally from performances and cannot be similarly tied to evoking emotions.

I propose a final thought-experiment that hopefully clarifies my position. Consider a conductor so adept at reading scores that he could do so at the piece?s tempo without missing any details(9). One might say this example conflates my distinction between the fictional machine (and in general, those who examine a score for its intrinsic properties) and concert-going listeners, since presumably a musician of such skill ?hears? an ensemble in his mind with the potential to move him as the real thing can an audience. Here I note again that the crucial detail possessed by our conductor and lacking in any machine (and by extension, any fixed work) is the progression by which his emotions can be induced. His journey through the piece is analogous to an audience?s without issue because it takes a comparable path from one measure to the next while his mind works in parallel. Claiming that the expressive properties thus experienced can transcend a ?performance? ? as I consider the example to be ? is not misleading merely because we lack a practical way to ?measure? them from a given work; it is a category mistake.

Appendix:
I noted earlier that I consider questions of emotions as depicted (but not aroused) by music to be tied to representational theories rather than the issues of expression I discussed. As for the obvious follow-up question ? to what extent I believe representational qualities are verifiable in a work ? I take more or less the compromise view, which I?ll try to explain briefly. I?m fond of Sparshott?s view of music as dialectic: in this context, referential meanings are evoked by the interplay between the composer & performer and an audience engaged in ?listening for.? Clearly this notion allows for representational variation within a single work: a performer can fail in his communication, or have an inattentive audience listening without direction. As he says, I call music sad (in the representational sense!) when it sounds like a sad person; this representational property could be turned 180 degrees by performance in a culture that considered sighing gestures (or whatever musical devices were used) the sign of a happy person.

Nevertheless, since we?re no longer applying the label ?sad music? necessarily to the emotions it rouses, a measure of objectivity is not without warrant. A modified black box that predicts what representational qualities an audience might perceive would be adequate for determining to what extent various qualities are inherent in the work. Some scientists would even call it a good model for the mind, claiming that humans as well as machines learn cultural norms by assimilating countless little rules corresponding to those entered by our machine?s programmer. More importantly, we are justified in maintaining that such intrinsic properties make sense because to some extent they retain their meaning when removed from any performance or even from the work itself. While I demonstrated earlier how the emotions experienced during an excerpt depend entirely on what is played before and after, there is no contradiction when, say, Wagner?s leitmotivs are listed without context in the program notes as independently signifying places, people, ideas, and yes ? emotions.

3. While the advent of recordings as the primary medium of consumption has brought many changes to our perception of music, I don?t feel it greatly affects the core philosophical issues. I think it?s clear enough that listening to a recording does little to change the way music can represent ideas and express emotions ? as with the final example of #2, there seems no harm in broadening the definition of ?performance? whenever the experience retains the sense of movement characterizing more traditional modes of listening. With regard to these roles in particular I?ll examine a few advantages and disadvantages, though I feel the former outweigh the latter. The phenomenon does raise some stickier questions regarding the ontology of music, but ultimately can be shown to fit within the model as proposed by Sparshott.
At first glance recordings seem to incorporate the whole expressive palette of ordinary music while adding unique capabilities. One is hard-pressed to think of a standalone musical event that cannot be recorded, but countless other techniques depend on it: multiple takes, overdubbing, effects, etc. A vocalist harmonizing with herself seems ordinary enough in our technological society, but was beyond imagination not so many years ago. Who knows exactly how today?s artists and listeners benefit from such methods, but there seems little doubt they have expanded the gamut of expressivity.

Or have they? I have to agree that the sheer variety possible in modern artistry implies that we?ve produced more ways we can move people. More generally, the depiction of thoughts is given great power: with the ability to record a bird instead of imitating it in the flute, for instance, the theories of artistic representation long applied to painting, sculpture, and so on can finally be brought to bear. However, we must not confuse representation with reproduction. Some ?audiophiles? are wont to spend a fortune trying to manufacture a way for their recordings to reproduce live music, but I take it as a truism that this task is impossible. Even if the engineering constraints were removed ? imagine, for example, a machine that replayed a performance directly into one?s brain ? the unique features of a recording would remain along with their philosophical implications (discussed later).

Perhaps the most problematic aspect of recordings is the way they blur our usual distinctions between composer & musician, between work & performance. Where do the studio engineers, editors, producers, and so forth fit? This one isn?t too difficult: they are performers, being equally charged with interpreting a work that is not theirs. Nor does a self-recording performer confound things; he merely plays more than one role, the same way a traditional musician does who happens to compose his own music. (I should emphasize moreover that the roles are completely separate: an artist performing his composition faces the exact same issues as does any other performer of that work, since by our model the task of interpretation is not, say, to determine the ?Urklang? held in mind by the composer, but is creative in its own way(10).) If anything, producers need more musicality(11) than other performers because they generally has the least direct contact with the original artistic vision, the fewest musical tools that actually alter a performance, yet hold much of the responsibility for the aesthetic success of the performance.

The ontology of this ?performance,? however, is not so clear. On the one hand, it certainly functions as a performance to consumers, who listen to recordings in virtually the same way they would a live act. In addition, as touched on earlier, they perceive representational and expressive properties similarly. (We may say at least that discrepancies specific to recordings, such as playback on a different system, vary the experience no more than factors like changing moods do to live performances.) However, a recording is fixed in an essential way that performances are not, a property we found to have great consequence with regard to musical works. Must we conclude, for example, that recordings cannot express emotion according to my logic in #2? I think not; we need to examine more closely.

A musical work as fashioned by a composer has no purpose, per se, except that which aestheticians ascribe; it is static by its very nature. In what sense is a recording static? Clearly its purpose is to preserve something that has been created by the various composers, performers, engineers, etc. involved. When we observe that a recording is therefore a means of manipulating the artists? product (which as described earlier is a kind of performance), not a creative product itself in the truest sense, we see that comparing it with a musical work is a kind of category mistake. Nevertheless this observation doesn?t by itself shed new light on the matter, for we can neither draw perfect analogies with any other previously-examined musical phenomena. The attempts are fruitful, though:

We?d probably try to compare with recordings with scores next. But recalling my example of an ideal conductor, I almost immediately see a continuum in information-theoretic terms: to generate a performance or its closest mental equivalent, a score requires assimilating lots of information but does not uniquely determine questions of interpretation; a recording consists of even more information but its ?playback? can be achieved with considerably less effort(12). An unrealized figured bass or a jazz chart might be even further removed than a score proper, requiring a measure of creativity beyond interpretation toward outright improvisation; a piano roll (or its computer-age equivalent, the MIDI file) is somewhere in between, very nearly a recording, but requiring an external instrument of the sort I?m reluctant to equate with amplifiers and speakers.

Returning the nature of recordings, we see now precisely in what sense we were forced to call recordings static: they are near the side of a spectrum requiring the least additional creativity to translate from some medium into sound; rather independently, we remark that their medium has the property of self-preservation. However, consider what lies at the end of this spectrum: performance itself! Once a live musician strikes a note, it is fixed in exactly the same sense, the only remaining determinants of sound for the audience being their choice of seat. Meanwhile, it obviously still possesses the sort of dynamic motion we require. We must not mistake the quick fading of physical sound waves for an ontological property.

In the light of the previous considerations, I find that the most probable ill effect of having recordings prevalent is not their inadequacies with regard to performance, but the way they have displaced the literacy of sheet music to solely musical and academic circles. If anything, our spectrum shows how the loss of this admittedly brief tradition might be comparable to a rise of live performances at the expense of recordings. We?ve seen how the use of a score demands (but no doubt rewards) a much more engaged ?audience,? in the end requiring that we postulate a fictional score-reader since nobody is as musically fluent with their eyes as with their ears.

To be sure, the time when scores predominated as the mode of distribution was a relatively short bubble between the Enlightenment and the Recording Age, but this point in history saw the formation of customs which no doubt enhanced many people?s appreciation of music. The French entertained with Chopin in their parlor-rooms; the Italians brought scores to watch Verdi. (When electric lighting first enabled directors to darken the house for dramatic effect, the operagoers rioted ? how were they supposed to follow the libretto?) Today?s environment, on the other hand, encourages our society to merely buy the CD, robbing people of the creative opportunities given (well, obligated) by a score.

The remaining objection for seeing recordings as equally communicative as performances is the apparent breakdown of the interplay between composer, performer, and listener. Our model for these relationships has more or less been Sparshott?s: a conversation. How can dialogue ever occur when musician and audience are removed in both time and space? I think this argument ignores the utility of the analogy while attempting to apply the terms outside their allotted scope (appealing to our common sense of language, in other words, instead of engaging the underlying ideas.) Viewing music as a shared activity, for instance, helps re-frame the problem of a work?s identity by acknowledging that participants can perceive it differently without destroying their communion, while casting doubt that there can be rules governing in a strict sense the way an artist must respond to a score any more than a conversational response is subject to rules. In no way do time & space affect such notions. After all, we have no trouble maintaining that performers can continue to interpret the works of a dead composer; I see no reason for the link between performers and listeners, however necessary, to be any more concrete.

Clearly then, there remains no reason to be hesitant about a recording?s ability to move people the way live music does. Its more trivial properties ? reproducibility, portability, and so on ? only seem to act as advantages for the musically inclined consumer who might otherwise never hear the Vienna Philharmonic, much less in his car and on demand. No doubt many now take it for granted, trivializing great works into ?background music,? but there is no actual loss here, only opportunity. Even allowing that many listeners require the direct interaction or unique sound or another aspect offered only by live performance, neither is there dichotomy here. Musicians have not stopped giving concerts the old-fashioned way.

(1)Certainly Aristotle didn?t mean only sensory pain, lest the definition be far too limiting. I?m all for a physiological connection ? yet once removed from directly perceived senses, what are pain & pleasure but emotions themselves? Even granting that they are somehow more fundamental, the claim would still be tautology: categorizing brain-states as ?emotional? or as ?containing pain/pleasure? would not strike me as coincidental (much less meaningful), merely as different labels for the same activity.

Meyer?s view that emotions are only pleasant or unpleasant in their cognitive context also seems relevant. Differences in cultural, situational, and personal beliefs strike me as the true determinants of pain & pleasure with regard to emotion, demonstrated e.g. by his example of a person falling: without knowledge of the outcome -> horrifying; at an amusement park -> thrilling. The origin of emotions (independent from their impact) is beyond the scope of this essay, however.

(2)Note one: I exclude the discussion of how music is moving if only because it?s such an ancient & protracted debate that it would take away from the central points of this essay (which has already required significant sidetracking). My thoughts on the matter will likely turn up in question 2, though.

Note two: I don?t want to imply the converse ? that because I link music and emotions, I believe language is best suited to expressing the most general of my overlapping categories, thoughts. If I had to make a similar claim, I?d choose feelings as the best match; the conventions of language like grammar are appropriate for this more definite class, while the former includes ideas (e.g., pure math, or even music itself) not readily expressible in any language. If you accept my definition of feelings as thoughts that influence our decisions, the conclusion is almost too neat & tidy: humans invented language to deal with thoughts that mattered in a day-to-day sense; the utility of abstract thinking only became apparent later (and with it, non-linguistic notations).

(3)If anything, I was tempted to argue that considering decisions based on emotions to be ?irrational? beyond the obvious fact that they?re not logical does disservice to our mind?s true capacity for reason; I find many points of agreement in Antonio Damasio?s theories.

(4)More precisely: the probability they would prefer one description to another.

(5)Hopefully I do so without dredging up the strict cognitivist point of view (and inevitably, its impassioned opponents). I don?t concern myself with whether the emotions represented in a work are themselves what stir the listener, or if its innate beauty does the trick; certainly that debate can go nowhere without placing claims on the internal experiences of others. (That Kivy is apparently wont to do so, despite grave misgivings with psychology proper, is one of many frustrations I had with his article despite accepting most of its thesis.)

(6)I?ll treat this issue briefly in an appendix.

(7)I realize a full refutation takes more than my reservations in footnote #5, or the deceivingly simple examples one is tempted to cite like ?I get pleasure from listening to melancholy works, ergo?? Nonetheless I think the utility of the distinction I drew should be evident even to those who are moved to actual sadness by Chopin et al.

(8)I?m reminded by this observation of Berleant?s description of repetition: ?regeneration rather than reiteration.? From the point of view I?ve expressed, it?s obvious why this characterization fits: when we listen to a figure for the second time, we?re not the same as we were before.

(9)Or in any case, none more than he would by hearing it performed by someone conducting an ensemble in exactly the same manner.

(10)While I?ll leave this assertion relatively undefended, I?ll counter the most obvious objection ? that performers have a set score and composers do not. While the two arts do have many differences, this matter is only a question of degree; all composers work within predetermined structures of some kind, even contemporary ones who probably think they don?t. More importantly, the kinds of choices made by performers are balancing acts toward an aesthetic goal, e.g. a virtuoso?s risk-taking vs. appearing effortless (see Grossman for countless more examples), of the exact same sort I feel underlie composition.

(11)Since this is among the hardest words to define in philosophy of music, let me explicitly quote Sparshott: ?a quality of understanding comparable to that necessary to composing a musically satisfactory piece.?

(12)No human I know of can read the grooved waveforms off an LP (or more unnatural still, the pits of a CD) in a manner approaching our ideal conductor and his score ? much less a listener and a real orchestra ? but I see this as a relatively minor consequence of the way we are wired to perceive and process different forms of information, not necessarily an indicator of its content or extent. Psychologists say we have a good deal of special-purpose hardware for handling aural stimuli, for instance, but doubtless there is no equivalent for reading digital bitstreams. I suggest as an apples-to-apples comparison the complexity required of a machine to interpret a score and reproduce a recording; primitive DSP chips accomplished the latter 20 years ago, but the former is far beyond our current reach.



Back to PhilosophyMusic