In this week's episode, Christopher Doll challenges our musical beliefs and aural skills training as he asks the question: 'Is Key Real?'".

This episode was produced by Megan Lyons.

SMT Pod's theme music was written by Zhangcheng Lu with closing music by David Voss. For supplementary materials on this episode and more information on our authors and composers, check out our website: https://smt-pod.org/episodes/season01/.

[SMT-Pod theme music playing]

Welcome to SMT-Pod, the premiere audio publication of the Society for Music Theory. In this episode, Christopher Doll challenges our musical beliefs and aural skills training as he asks the question: 'Is Key Real?'"

Key is one of the most fundamental concepts in all of Western music theory. But is it real? This question takes its inspiration from a 1992 essay by Harold Powers called “Is Mode Real?,” although I want to approach key a bit differently from how Powers approaches mode. When I ask if “key is real,” I mean whether it’s aurally real—that is, whether it’s something to be experienced, not merely something we talk about. I think it’s safe to assume that most Western musicians, whether formally trained or self-taught, assume key is real. Compositions by classical composers often indicate the key in their titles. And some compositions like Bach’s Well Tempered Clavier [music] or Chopin’s Op. 28 piano Preludes [music] are methodically organized around the 12 major and 12 minor keys. Yet acknowledging these facts doesn’t actually answer the question of whether key is something we can hear. It’s obvious that key is real as an idea; it’s less obvious that key is real as an experience.

There are actually various distinct terms we use to denote the supposed experience of hearing the emergence of a key, terms like “key finding” or “tonality induction.” But again, the mere fact that these terms exist isn’t sufficient to answer the philosophical question I’m pursuing. The sheer presence of expressions like “key finding” and “tonality induction” doesn’t suggest the reality of the experience in question any more than the existence of the words “witchcraft” or “sorcery” suggest the reality of magic.

In asking whether key is aurally real, I’m trying to frame the listening experience in a particular way, so that it’s distinct, if not entirely separable, from intellection. There are longstanding debates in psychological and philosophical and music-theoretical literature about the relationship between perception and cognition, and between experience and thought, and I won’t attempt to enter that debate here. But I do want to insulate experience in such a way that helps us avoid wandering into obviously intellectual terrain. For instance, the notion of monotonality I consider largely an intellectual concern. In short pieces or songs, I find monotonality entirely plausible as a description of a possible aural experience. But in longer and more tonally complex pieces, where the concept really achieves its explanatory power, monotonality, I would argue, even for the most highly trained listeners with absolute pitch, moves into the realm of conjecture and ultimately of fantasy.

There’s nothing wrong with fantasy. There’s nothing wrong with conjecture or intellection more generally. I simply wish to distinguish intellectual matters from experiential matters as much as possible. Now, there’s overlap between experience and thought that’s unavoidable. And there’s a compelling argument to be made, one articulated by psychiatrist Iain McGilchrist among others, that talking about (or even thinking about) any experience is by definition to surrender to a left-brain fictionalization of what’s really going on. Without pretending to have escaped these conundrums, I just want to make clear that my discussion will try to focus exclusively on matters of aural experience, and I’m being fussy on this point because adherence to aural reality isn’t built into many of the standard music-theory terms on which I’ll need to rely in my discussion. By the way, I identify a number of relevant scholarly sources in the bibliography for this podcast. This is a big topic, so the bibliography just gives a glimpse of some of the relevant work on this issue.

Now, even among the myriad musicians who take the aural reality of key as almost self-evident, there are competing ideas about which musical features give rise to the experience of key. One such idea, found in the work of David Butler, is that when we encounter intervals that are relatively rare in a major or minor scale, like the tritone or the semitone, we arrange the intervals in our minds to fit within one of those scales, and this corresponds to a particular key. Another idea, one important in my own work but that has precedents going back at least to the Medieval notion of the modal final, is that listeners recognize melodic or harmonic formulas that may or may not contain any rare intervals per se, but that fit into a certain scale in a specific way, and this corresponds to a particular key.

Yet another idea, prominent in the writings of Carol Krumhansl and many other authors attached to or following her work, is that emphases on certain pitch classes, especially emphases from greater duration and frequency of occurrence, but also possibly from stronger metric or hypermetric placement, relative loudness, and other parametric disparities, results in our hearing these pitch classes as hierarchically important, and this hierarchy corresponds to a particular key.

It’s also possible, and I think likely, that these and other key-finding methods are in play to different degrees at different times. In a composition, for example, like Terry Riley’s minimalist classic In C from 1964, we hear, at the beginning, a rearticulated drone of high octave Cs [music] and a melodic riff of C and E [music]. There are no rare intervals here, nor any key-defining melodic or harmonic formulas beyond the octave and major third. The relative duration and frequency of occurrence of the Cs seem like the primary reasons to hear the opening of In C in C, if we’re to hear a key at all.

Now contrast this case with your typical rendition of “Happy Birthday.” [music] The tonic note arrives only intermittently and never hypermetrically accented. As we wait impatiently to blow out our candles, we’re presumably hearing one key or another, or maybe multiple ones simultaneously depending on the abilities of the singers [Happy Birthday plays]. Even though there’s not much emphasis on the tonic note in “Happy Birthday,” aside from being held a tad longer, we’re able to plug the tune into our tonal memories and, through some means, experience something we call “key.” But the key in the case really doesn’t depend on the duration, metric placement, or frequency of occurrence of the tonic note.

My point here is that we don’t yet have anything like a complete understanding of the mechanics of hearing a key. A major problem here is in the idea of key itself—what scholar Piet Vos identifies as the fuzziness of the underlying concept. One way to deal with this fuzziness is simply to start over; new term, new idea. But it’s hard to accept that hearing key is simply a figment of our collective musical imagination. There’s obviously something—or multiple things—about the idea that many musicians find experientially evocative. I’m of the opinion that key is real—at least real in part. I propose splitting key into four parts—tonal center, the tonic triad, the scale or mode, and functional harmony. When we use the word “key,” we’re often talking about some combination of these four things, and by breaking down the concept in this way, I’ll be better able to distinguish which aspects of key correspond to something we can actually hear.

[bumper music]

The first element, tonal center, I define as the quality of maximum stability surrounding an individual pitch class. Various pitch classes might sound stable; but relative to all others, the centric pitch class is the most stable of all. I’m not actually a fan of the metaphor of centricity itself, because when I experience tonal center, it feels to me less like focusing on the middle of a circle or locating an object’s center of mass, and more like orientating myself in space. The energetics are such that when I hear centric ambiguity—that is, two or more centers simultaneously—it’s like trying to face multiple directions at the same time. However, tonal center as a term is very useful rhetorically, because center and centricity don’t have a zillion other meanings like so many music-theory terms do. The word “tonic,” for instance, can be a synonym for tonal center, but it can also denote a triad built on the tonal center, or it can denote the function of various chords (as when talk about roman numeral vi or iii functioning as tonic substitutes). So “tonal center” is my term for scale-degree ^1.

Now, is tonal center aurally real? Yes, I would say so. Tonal center is probably the most obviously experiential aspect of key, which ironically can make it a site of great contention among listeners when they’re comparing aural results. As someone invested in researching the tonality of popular music, I can’t count the number of times people have come to me with dissenting opinions about the tonal center of this or that song. Differences in opinion oftentimes I think are rooted in the centric cues of the music itself, which is to say, I think popular music especially (though certainly not exclusively) tends to feature assorted kinds of ambiguities with regard to tonal center, and this lends the music to competing experiences. This experiential multiplicity can be observed not only between different listeners, but also within the experiences of individuals who are attuned to certain musical features in certain ways.

As a brief illustration, consider the vamp from Lili Allen’s hit song “Smile” from 2006. Her chords are sampled from a 1966 Jamaican rocksteady song called “Free Soul” by the Soul Brothers, but it’s a simple oscillation very much in the tradition of folk music like the sea shanty “Drunken Sailor.” [music] In “Free Soul” and in “Smile,” the chords have some extensions, but essentially offer the same back-and-forth motion, on G and F [music]. Since the G chord is the first one we hear, and it’s hypermetrically accented, and it’s part of formulaic pattern we might already know, I would guess that a good number of us experience G as the tonal center at the start of the track. Yet as the harmonic repetition continues, I hear the tonal center shift, to F.

The pentatonic vocal fragments of D-C-A in the verse and in the pre-chorus suggest center F more than center G to my ear, and I find the vocal line in the chorus especially persuasive, as Allen arpeggiates an FM triad over both chords [music]. In my own writings, I’ve called this experience of shifting from one center to another while the notes remain the same the “reorientation effect.” At any rate, I think we’re justified in making the larger claim that the concept of tonal center in general describes something aurally experiential—the sensation that there’s a sort of tonal gravity that the pitches are obeying. Tonal center is real.

Before moving on to the other three elements of key, I want to make it clear that when I say something is aurally real I don’t mean to imply that it’s also necessary, or good, or natural, or even objective in the way other sonic objects like a sound wave are objective. Tonal center is real, and objective, only to the degree we can surmise such hearings by asking listeners veiled questions about how well a certain tone fits with a given audio sample, or to the degree that listeners self-report hearing tonal centers. I’m one such self-reporting listener; presumably some of you hearing this podcast are as well. But take another listener, even a highly trained one, from a different musical tradition, and there’s no guarantee we’ll get the same results from our aural reality check. Tonal center certainly may not map onto every listener’s experience in an obvious way, but there’s enough intersubjective evidence to suggest that it’s real for at least some of us, some of the time. That’s good enough for me.

Now on to the second element of key, the tonic triad. You may be wondering why I bother to distinguish between the tonic triad and tonal center; after all, they both go by the same term “tonic,” as I mentioned. In truth, the situation is even more terminologically complicated than I earlier admitted. I mentioned that the word “tonic” can describe a pitch class, or a triad, or a harmonic function. But when it describes a triad there’s yet another distinction to be made, which is between the tonic and a tonic triad. The tonic triad is an abstract entity akin to tonal center, a force of gravity that represents maximum stability (just with three notes instead of one). A tonic triad is a particular chord in a piece of music, often represented by a roman numeral I. Regardless of whether a tonic triad, a I chord, appears in a given passage, the assumption is that the tonic triad for that passage is still in effect. If a tonic triad is present, we’ll hear it. But what about the tonic triad: is this something we can experience?

To answer this tricky question, I think it’s important first to look more closely at the distinction between the tonic triad and tonal center. Tonal center makes sense to me on an experiential level, as a sensation of maximum stability on a single point in pitch-class space. The tonic triad distributes that maximum stability across three pitch classes, in a way that I’m not sure is quite coherent. Are the fifth and third of the tonic triad less stable than the root? Yes, generally they’re theorized that way, and there’s experimental evidence to suggest we hear them that way. So the third and fifth are not actually maximally stable; they’re still subordinate to their chordal root, the tonal center. Are the third and fifth equally less stable? I don’t know, and experimental evidence isn’t clear on this point.

Are the third and fifth less stable than the tonal center but more stable than other notes in the key? Maybe, sometimes. Experimental evidence certainly suggests that overall scale-degrees 3 and 5 sound more stable than scale-degrees 2, 4, 6, and 7, but of course, this is totally context-dependent. It’s one thing to hear 3 and 5 as relatively stable at the end of a passage, where we can hear (or imagine) a tonic sonority providing resolution. But if those notes appear as part of, say, a stepwise bass line ascending through scale-degrees 3, 4, and 5 at a cadence, the effect is quite different; those notes are highly unstable [music]. For the tonic triad to make sense, it must be abstract enough so as to not be affected by such local contextualizations. So then, what would it mean to hear such an abstract sonority?

Psychologist Richard Parncutt has argued not only for the aural reality of the tonic triad, but for the unreality of tonal center as separate from the tonic triad. He bases his claim in significant part on the tradition of Schenkerian music theory, which is indeed where I think the strongest theoretical answer to the question of the tonic triad’s abstraction can be found. Heinrich Schenker posited that the tonic triad constantly operates at a background level of a tonal piece of music. All the sounds of the musical surface—the note-to-note utterances—serve to prolong that background tonic triad. In this view, non-tonic-triadic notes do merely displace tonic-triadic notes, they also in some sense elaborate tonic-triadic notes that are still present at a deeper level, so even when none of the notes of the tonic triad are literally sounding, the tonic triad is still metaphorically sounding, mentally retained in a motionless state, as Schenker said.

This is a beautiful idea, but I’m at a loss when trying to connect this to actual listening. It’s one thing to claim that the compositional process often involves diminution, adding notes to predetermined, simpler musical structures. It’s another thing to say knowledge of this kind of process can inform how listeners hear notes in relation to one another. And it’s yet another thing to say that we actually hear that predetermined, simpler, underlying musical structure—not merely the results of its diminution, but the structure itself—all the while hearing its elaboration. I don’t even know what that means, exactly. This strikes me as a form of mysticism, less music theory than music theology, to paraphrase composer George Edwards. Prolongation of the tonic triad may or may not speak to some deeper musical truth, but from a purely experiential perspective, I don’t know how to make sense of this idea.

As for experimental evidence for the abstract tonic triad, we might look to studies I mentioned earlier that clearly indicate scale-degrees 1, 3, and 5 as fitting better, or sounding more stable, when listeners compare them to other notes. This finding may be consistent with the aural reality of the tonic triad, but it’s by no means evidence of its existence. The tonic triad is a triad—a sonority of three notes fused together—ostensibly operating at some abstract level. Knowing that listeners find scale-degrees 1, 3, and 5 to be more stable individually doesn’t mean those notes come as package, and it certainly doesn’t mean that package is constantly looming in background. Again, the gravitational weight of the root by itself, the tonal center, I understand. Where I have difficulty is in figuring out just which aspects of my aural experience are supposedly being expressed by the chordal third and fifth of the tonic triad.

Matters are even more perplexing when we encounter music that isn’t isn't straightforward with its third scale degree. Take a simple looping triadic progression with roots on scale-degrees 1, minor 3, and 4 [music]. This figure is common to sixties rhythm and blues and rock and roll, and many of their descendants. Sometimes, in such passages, the roman-numeral i chords are powerchords (containing a root and fifth but no third). They appear as powerchords in the 1962 R&B classic “Green Onions” by Booker T. and the M.G.’s [music]. Even though we don’t encounter a tonic chord anywhere in the song with a clear minor third, we might still be willing to say the tonic triad of the song is minor, given the prominence of the second chord in the series, which is rooted in the scale’s minor third. But other times, the quality of the tonic triad isn’t so easy to divine.

In the choruses of Beatles’ “Magical Mystery Tour” from 1967, every individual tonic chord is an E major triad, followed by GM and AM [music]. And in one pass, there’s even a chordal minor seventh added to the E major triad in Paul McCartney’s vocal line, ostensibly creating a dominant seventh chord. There’s a pretty big difference between a tonic dominant seventh chord and the tonic minor triad we derived in “Green Onions.” Do we think that, despite this discrepancy, the Beatles passage likewise has a minor triad as the tonic triad? What about in another 1967 rock classic, “Purple Haze” by the Jimi Hendrix Experience [music]? Here, all the roman-numeral I chords are instances of the famous “Hendrix chord” often written as dominant 7 sharp 9, but which also can be understood as a major triad and minor triad combined, along with a minor seventh. Is this overly full psychedelic concoction the tonic triad? Clearly, the tonic triad can’t in every case be derived straightforwardly from a tonic chord. So, where does that leave us?

Well, some of us might say these songs aren’t in a key at all. I doubt many of the musicians who created the songs would say that, and I wouldn’t say that, but it’s possibility. Alternatively, we could instate some rules about how such problematic cases are to be resolved—for instance, we might look to the scale to which the roots of a chord progression belong for determining the quality of the tonic triad, which would mean all the examples I just mentioned would have a minor triad for the tonic chord. But following this logic, in “Magical Mystery Tour,” the tonic triad would be minor, even though every I chord we actually hear has a chordal major third [music].

We might also think about expanding the tonic triad to tonic polychords like Robert Bailey’s double-tonic complex, or to non-triadic tonic chords, as has long been suggested by theorists working on highly chromatic and post-tonal music. Or we might contract the tonic triad to just a powerchord—a root and a fifth. Either way, expanding or contracting, I still don’t see a clear bridge between the abstract sonority and aural experience. I can’t even suggest an experiment to disprove the notion of the tonic triad, or the tonic chord, as an experiential phenomenon. In a word, the idea seems to me unfalsifiable. That’s a problem. Unfalsifiable things are not things we need to take seriously. All this is to say, I believe that this component of key, the tonic triad, is not aurally real [music].

This brings me to the third of the four elements of key. In this context, I see no meaningful distinction between scale and mode, so I’ll just say “scale” hereafter. Scale is one of the two explicit aspects of the taxonomy of keys, the other being tonal center. Key signatures indicate nothing but scale, or more accurately, they indicate the accidentals for the diatonic notes of the scale, although of course for minor this is complicated regarding scale degrees 6 and 7. But at any rate, scale is undeniably a major component of key. But is scale real?

Yes, but it’s not that simple. There’s something similar going on with scale as there was with tonic triads. Just like there was a tonic and the tonic, there’s a scale and the scale. The opening line of the Lowell Mason version of “Joy to the World” is a scale, a descending one at that [music]. But there’s also the scale of “Joy to the World,” the one it shares with “Jingle Bells” and “Silent Night” and every other major-based Christmas carol. “Jingle Bells” and “Silent Night” don’t start with a scale, yet the scale of both tunes is obvious—it’s the major diatonic. When we look at minor, matters are more slippery. Nowadays, musicians tend to talk about three standard minor scales—natural, harmonic, and melodic—and these are perfectly good ideas when it comes to describing a scale.

But what is the scale of a minor key? I don’t know, and I don’t think anyone else does either. Some theorists have floated ideas about pitch-class collections that are larger, or potentially larger, than a diatonic scale, that have a cardinality greater than seven unique notes. Such collections, like David Temperley’s “supermode” or Dmitri Tymoczko’s “macroharmony,” can subsume the different versions of scale-degrees 6 and 7, or even fold all three versions of the minor heptatonic scale together with the major diatonic. These are compelling ideas, but for our purposes here, it suffices to say that a scale is distinct from the scale when talking about the components of key. The scale is a centered collection of pitch-classes; a scale is a stepwise melodic line of pitches.

A scale is quite obviously an object of experience. Anyone who’s ever seriously practiced singing or playing an instrument knows all too well the experience of hearing a scale—over, and over again. The scale, however, the more abstract entity, the one that is an element of key, still needs more unpacking to evaluate aurally. The notion of the scale is that there’s a set of notes arranged around a tonal center that are primary, that are the default notes, that are in-group. It seems to me that if we are to consider the scale as a description of experience, the best place to observe such an experience is at its edges—that is, by hunting for notes that are out-group.

The standard term for in-group notes is “diatonic”; out-group notes are “chromatic.” Can we hear chromatic notes? I would say yes, but there are at least two different potential experiences available here. Sometimes we say notes outside the key sound outside; a foreign note substitutes for what could have been a native note. The concept of modal mixture or modal borrowing operate on this notion; you have your in-group notes and then you insert a foreign note that comes from a different scale, like the venerable Picardy third, an intrusion of major into an otherwise minor key. A cruder version of the same experience can be achieved by playing any tonal piece of music and then having your toddler improvise embellishments [music]. This is definitely something we can hear.

A different kind of experience is implied when we say chromatic notes sound like altered versions of diatonic notes. In this case, there’s no entirely foreign note, just a native note that has transformed like a caterpillar into a butterfly, a diatonic note that has been chromaticized. This line of thinking is reflected in our practice of adding accidentals to seven unchanging diatonic numbers: natural scale-degree 2 has become flat-2 [music]. Changing a thing of course isn’t at all the same as exchanging that thing, and it implies a distinct attendant experience. But I can safely say that I hear scale-degrees chromatically transform all the time. So I find both experiential versions of the scale to be real.

However, I already said matters weren’t that simple. That’s because the scale isn’t always real, at least not how it’s traditionally thought of. For one thing, keys are conventionally theorized in the context of the major-minor system, and such a system clearly isn’t adequate for describing the tonal properties of a lot of Western tonal music from the last few centuries. Any tune consistently in the dorian or mixolydian mode (or scale, as I prefer), is going to resist categorization into a major or minor key. That also goes for the rarer phrygian, lydian, and locrian, as well as other seven-note scales like harmonic major, or the double harmonic major famously used in the folk song “Misirlou,” or more recently in the Black-Eyed Peas 2005 hip-hop song “Pump It,” which samples the surf-guitar rendition of “Misirlou” by Dick Dale and the Del-Tones [music].

Some listeners might hear all these other scales as alterations of major and minor, and I won’t deny that’s an experiential possibility. But it’s also experientially possible to hear those colorful flat-2s as in-group notes, especially when they’re the only version of the scale-degree sounded, in which case the scale can be considered a description of experience only if the list of potential scales for keys is expanded to include things like phrygian or double harmonic major.

Now, let’s say we’re willing to concede that point. We’re willing to say “Misirlou” sounds in the key of E double harmonic major. The experiential problems with the scale of keys don’t end there. They continue as we home in on music that habitually mixes major diatonic and minor pentatonic scales. Consider this outline of a blues riff [music]. This is very similar to the progressions of three chords I discussed earlier, just in a slightly different order, and usually played over one longer-lasting chord rooted in the lowest note [music]. We hear versions of this riff in many songs, among them Bo Diddley’s “I’m a Man” and Muddy Waters’ response to Diddley, “Manish Boy,” both from 1955. The riff’s bass line adheres to the minor pentatonic scale [music], but the upper line suggests major—either major diatonic or major pentatonic [music].

In thoroughly mixed-scale music like this, I often don’t hear notes projecting strong in-group versus out-group identities, or if they do, these distinctions involve multiple different scales simultaneously. They’re all tangled in a poly-scalar ball, and I don’t find it experientially honest to declare one those scales the scale. Maybe it’d make sense to start talking about multiple simultaneous keys in these and other cases, but I’m not convinced the scale is an aurally real thing here. That’s why I said the reality of the scale isn’t so simple. I find it’s real some of the time, but not all of the time. And in those cases where I can't hear it, I think it’s appropriate to take an inclusive attitude and embrace scales beyond the major and minor diatonic. [music]

And this brings us to the last of the four components of key, functional harmony. This component is unique in that it’s not directly observable in the name of a specific key. And its contents aren’t based on a specific key. The tonal center, the tonic triad, and the scale of a piece of music in C Major are different from those of a piece in D Minor. But the functional harmony of both pieces, and every piece, is always the same. Functional harmony is just functional harmony, regardless of the particular key. I must say, I dislike the expression “functional harmony,” because it’s led to the ugly and unhelpful expression “non-functional harmony,” which is a little like saying “non-classical music.”

But at any rate, the concept of functional harmony as it relates to key is that a piece of music is often thought to be in a key only if it’s confirmed through statements of a handful of cadential chord progressions dating from the Baroque era. Paramount is the inclusion a roman number V moving to I, with the leading-tone resolving to the tonal center. No matter how recent the music is, this is still frequently the standard that theorists use to decide whether a key is present, and especially whether the key has changed. From this perspective, most of the music I’ve cited in this podcast wouldn’t cut the mustard. Those pieces don’t feature functional harmony of this sort, and therefore they wouldn’t be in a key at all.

We’re again confronted here with a blend of intellectual and experiential concerns. It can simply be a matter of definition that key must be confirmed through a V to I (or so-called authentic) cadence. That’s a definition we can choose to follow and enforce, or not. Sometimes, theorists forego the authentic cadence per se and merely stipulate there must be some kind of V to I progression somewhere (which is why I’m calling this topic “functional harmony”). I’m not so interested in these kinds of rules for their own sake, though. What interests me more is the experiential side. Whatever we call it, the key of a piece of music is without a doubt strengthened when confirmed by an authentic cadence or a V to I progression—that is, by so-called functional harmony. This is especially obvious when composing a modulation from one key to another. But to my mind, the requirement of functional harmony for aural recognition of a key seems to me misplaced.

From an aural perspective, functional harmony is more appropriately considered alongside rare intervals, and melodic/harmonic formulas, and pitch-class emphasis through duration and frequency of occurrence. That is, functional harmony should be placed together with those musical features theorized as potential key-finding cues. And as I said a while back, we don’t yet have a complete understanding of how key-finding works as an aural phenomenon. But I’m prepared to state the no single one of those key-finding methods is absolutely essential to hearing key, or at least to hearing tonal center and the scale. If V-I progressions were truly requisite, then a simple round like “Row, Row, Row, Your Boat” wouldn’t be in a key, because it never leaves its I chord. That’s a strange claim to me [music].

So, is functional harmony real? Yes, definitely. But I would argue that it its relationship to key is made more coherent if we recast it as yet another potential key-finding method, rather than considering it a defining element of key. Functional harmony, I’d say, is real but often irrelevant, and not in the same experiential category as the tonal center and the scale. [music]

Music theory is always a theory of some music—some music and not other music. Theories of sonata form are very useful for engaging eighteenth and nineteenth-century European symphonies and sonatas and string quartets, engaging them both intellectually and aurally. Theories of key are very useful for engaging music that unambiguously exhibits the Baroque major-minor system. The cracks in key begin to show when applied elsewhere, which might lead us to circumscribe our application of the term. No doubt, this is a primary motivation behind the supposed requirement of functional harmony. It’s a way for a music theory to identify its relevant repertory. If there’s no V–I, then key doesn’t apply. But as I said near the opening of the podcast, the term “key” is used by musicians in all sorts of Western traditions, not just ones that are unequivocally based in the major-minor system.

There are good reasons to craft music theories carefully and narrowly, but there are also good reasons to take seriously what musicians say about their own work. These competing values can be balanced in different ways, and I’m more than willing to accept that my own opinions on this balance as regards key aren’t for everyone. But if making music theory evocative of aural experience is also something we value, which I believe we should, then it’s appropriate to subject our ideas to the occasional aural reality check, regardless of where we draw the lines around the repertory. Even in the C major preludes by Bach and Chopin, I don’t think the tonic triad is aurally real. And I think the functional harmony of these preludes, though real, isn’t necessary to the key; after all, I can hear the key just find from the first measures alone, with just the I chords. This is true even if I transpose them. This is B major [music]. This is D-flat major [music].

So, is key real? Yes, if by “key” we mean tonal center, and sometimes the scale. And functional harmony is real, but not in the way it’s sometimes said to be, as a requirement for hearing a key. Of course, these are claims about experience, and by nature everyone’s experience is unique. Perhaps some of you listening to this podcast are positive that you hear the tonic triad, and always hear the scale. I can’t claim to know otherwise. And if I’m indeed missing out on these experiences, I sincerely hope that changes soon. Thinking and talking about experience is a tricky business. To paraphrase the famous physicist Richard Feynman, our first principle should be to avoid fooling one’s self, because one’s self is the easiest person to fool. The threat of confirmation bias is everywhere and ever present. Yet this is still a good business to be in.

After all, the listening experience is presumably what motivated most of us to study music in the first place. And as the old tale tells us, as recounted by philosopher Abraham Kaplan, the drunkard who lost his ring of keys in the darkness is silly to search for them under the lamppost merely because the light there is better. Our fumbling in the dark, examining our experiences, is the best way to look for our keys, at least until future breakthroughs in neuroimaging or artificial intelligence can point us in new directions. For now, I’ll close with a brief anecdote about one of my graduate-school mentors, the late Jonathan Kramer, who once told a class, “You’ve probably all been taught the fiction that there are three kinds of minor scales.” A student then asked, “If that’s the fiction, what’s the reality?” Jonathan then responded: “There is no reality.” [music]

Many thanks to Zoe Doll, Rosie Doll, David Temperley, Jennifer Beavers, and Megan Lyons for making this episode possible.

Visit our website for supplemental materials related to this episode at smt-pod.org. And join in the conversation by tweeting us your questions and comments @SMT_Pod. SMT Pod's theme music was written by Zhangcheng Lu with closing music by David Voss. Thanks for listening!