Can we cool it a bit with the "note-spelling"?

[Adventure Temple Trail]

In recent years, it seems as though the term "score clue" has undergone semantic narrowing to the point that it almost always means "transcribed string of notes", and that kind of clue ("in this piece, the trumpet melody goes D-E-F-F-etc.") has also become significantly more common. I have sometimes been helped by these clues, and I believe there should be some sometimes. But I find it unfortunate that the "note spelling" subtype of score clue seems to be crowding out other kinds of clue in auditory fine arts questions.

I'd say that if you're doing more than one of these every 3-4 packets or so, it's probably overkill. I'd also warn people who wrote them to be extra careful about:

• Making sure they're transcribed in the right key
• Keeping them as short as possible; John Lawrence has previously argued for "no more than five notes" and that sounds like a decent rule of thumb to me
• How omitting features like rhythm and which octave the note is in can make it very hard to parse in real time (and trying to add them can make it clunky); something like "D-A-D" doesn't disambiguate "ascending D-A-high D", "D, up to A, down to D", etc.
• Using them only for the most auditorily famous or distinct parts of a composition (and being comfortable placing them in the middle-to-late clues where appropriate)

I'll also re-up this quote from a prior discussion in Discord, in case people are wary of note-spelling and unsure what other kinds of clue they can use:

There are dozens of ways to clue music, and there's nothing special about so-called "score clues" (meaning, there's no reason for them to be near the top of a hypothetical sorted list). An unsorted list of some music clue types: instrument mechanism · extended techniques · treatise · pedagogy · acoustics · pure theory · music notation · music publishers · music history · ethnomusicology · musicology/analysis · original analysis · arrangements/quotations · sounds similar · composition inception · composer quip · composer style · composer biography · composer relationships · musician biography · writings on composers · score clue · performance practice · performance/recording history · conductors · programmatic music · commonality (e.g. something depicted by music) · catalog numerology · term or coinage · literary basis

[L.H.O.O.Q.]

Probably one of my less popular quizbowl writing opinions is that whatever one's opinion is of note-listing score clues should also be one's opinion of computational clues. I would posit that the ability to "hear" a melody in your head based on a list of note names at game speed is similar to the ability to perform a requested calculation at game speed. I would also posit that, despite being difficult to play, score clues actually demand a relatively surface-level engagement with subject matter - akin to those annoying clues that just ask for minute, impossible-to-place details of a specific painting.

(I also just find note-listing clues aesthetically unappealing. They slow the tossup to a crawl and completely disrupt a moderator's flow!)

[cwasims]

In recent years, it seems as though the term "score clue" has undergone semantic narrowing to the point that it almost always means "transcribed string of notes", and that kind of clue ("in this piece, the trumpet melody goes D-E-F-F-etc.") has also become significantly more common. I have sometimes been helped by these clues, and I believe there should be some sometimes. But I find it unfortunate that the "note spelling" subtype of score clue seems to be crowding out other kinds of clue in auditory fine arts questions.

I'd say that if you're doing more than one of these every 3-4 packets or so, it's probably overkill. I'd also warn people who wrote them to be extra careful about:

• Making sure they're transcribed in the right key
• Keeping them as short as possible; John Lawrence has previously argued for "no more than five notes" and that sounds like a decent rule of thumb to me
• How omitting features like rhythm and which octave the note is in can make it very hard to parse in real time (and trying to add them can make it clunky); something like "D-A-D" doesn't disambiguate "ascending D-A-high D", "D, up to A, down to D", etc.
• Using them only for the most auditorily famous or distinct parts of a composition (and being comfortable placing them in the middle-to-late clues where appropriate)

I'll also re-up this quote from a prior discussion in Discord, in case people are wary of note-spelling and unsure what other kinds of clue they can use:

There are dozens of ways to clue music, and there's nothing special about so-called "score clues" (meaning, there's no reason for them to be near the top of a hypothetical sorted list). An unsorted list of some music clue types: instrument mechanism · extended techniques · treatise · pedagogy · acoustics · pure theory · music notation · music publishers · music history · ethnomusicology · musicology/analysis · original analysis · arrangements/quotations · sounds similar · composition inception · composer quip · composer style · composer biography · composer relationships · musician biography · writings on composers · score clue · performance practice · performance/recording history · conductors · programmatic music · commonality (e.g. something depicted by music) · catalog numerology · term or coinage · literary basis

My opinions on this topic are likely easily-inferred from various past music questions I've written for 2020 and 2021 ACF Winter and various iterations of WORKSHOP, but I don't agree with many parts of this post. First, I think there is a large gap between perceptions and reality of how common these clues are: it's rare for a tossup to have more than two note sequences and in most sets I imagine at least a quarter-third of tossups don't have any. Given that most note sequences are accompanied by some indication of musical expression, tempo, dynamics, etc., this means that a very small amount of the total text of music tossups is taken up by pure note sequences.

Second, I think that people who are not music specialists underestimate how helpful note sequence clues can be. Although I understand that they're less accessible than some types of music clues, I also think their inaccessibility is somewhat overstated: note sequences tend to be from fairly famous/recognizable sections of a piece of music, and understanding music notation, while not trivial, is not a particularly difficult skill to learn (certainly no more difficult than learning advanced science!) and is anyhow common among the QB-playing audience. People often refer to other kinds of "more accessible" score clues as being adequate substitutes for note sequences but I am not convinced this is necessarily true. For instance, I think score clues about instruments are often quite unhelpful (e.g. "in this symphony, the clarinets play a soft theme followed by a loud theme from the trumpets"). That's not to say we shouldn't try to more frequently clue aspects of the score that aren't note sequences when we can (I think my Beethoven piano sonatas tossup from 2021 ACF Winter did a good job with this), but just that note sequences are probably the most "reliable" type of score clue.

In terms of the mechanics of writing note sequence clues, I'm not sure that brevity is a virtue. I recall a few years ago playing a tossup on Scheherazade that used the note sequence "C-D-E, D sharp-E", and being unable to confidently buzz in part because it was too short for me to fully recognize it. However, if the score clue had continued with the rest of the phrase "G-F-E, D sharp-E", then I almost certainly would have converted it, because the full sequence of notes is far more evocative (at least to me) than just the partial one. Obviously, it's important to be judicious and not include long note sequences just because you can, but imposing a hard limit of five notes is far too restrictive and in many cases will likely hinder the conversion of these clues.

I generally agree with the rest of the notes about mechanics - if you're confused by transposing instruments, it can help to look up the same passage in a piano transcription so as not to get confused by which key you should be writing them in. In terms of rhythm, I try to put a comma after longer notes and dashes after shorter notes - I know this is not super ideal and may not always be correctly interpreted, but I think it's better than always putting "long" and "short" within the score clues themselves, especially when lack of explicit rhythmic indications is less problematic for parsing the particular note sequence.

I personally think that score clues are essential to any good set of music questions, appropriately balanced with the many other possible types of clues that exist. Score clues are one of the few ways to consistently reward listening to (or playing) music, which is obviously the way that music is meant to be consumed for the most part. It is especially difficult to clue pieces without strong programmatic elements or memorable backstories without using them, and these pieces already tend to be underrepresented in QB as-is.

[Eddie]

Hi everyone,

I have not regularly written or edited music questions in a long time, for I am old and do not matter much. Having said this, and with all due respect, I agree with Matt and Sarah, and I disagree with Chris. Much of what I have to say has already been said by those more wellspoken than myself, such as Ophir in the section "What are successful clues?" of the writing "Evasiveness in music tossups", and John in the section "Part III: Clues about the Text of the Piece Itself" of the forum post "A Detailed and Practical Guide To Music Writing/Editing". Moreover, I have made three attempts at writing here a coherent defence of my position, and have failed each time. Hence, I am instead giving only a short list of my hot takes, all without justification until pressed further.

A good note transcription clue has the following properties:

• It does not reward listening to music. This property has since been recanted.
• It may reward performing music.
• It rewards reading scores.
• It describes a piece that can otherwise be clued both (a) out of power and (b) without the use of note transcription.
• It describes either (a) the opening passage or (b) the main theme of the best-known movement.
• It immediately follows enough tempo/instrumentation context to reconstruct the passage with reasonable precision.
• It can be parsed and buzzed on at game speed by someone who has never played quizbowl.

[Subotai the Valiant, Final Dog of War]

I think the issue with all forms of score clues is mostly with badly written score clues that are too vague to be buzzed on. Being non-evasive and specific is definitely a requirement, and I agree with Chris that "note clues are the most 'reliable' type of score clue." I agree that parsing a note clue is akin to parsing scientific equations.

I am a fan of combining notes and other information as best as possible, since more useful information makes it more likely that any given person can buzz on the clue with the same amount of existing knowledge; in that sense, I'm not actually a fan of pure note clues. But I also don't think there are that many questions where pure strings of notes and rhythms are being listed without any additional context.

I think ultimately that in at least some music questions, one should be rewarded for directly engaging with the music. I'm not sure what means of rewarding players for knowing the music itself exist without describing the sounds in the music in some way; the most reliable way to do that is with score clues, and with as much specific detail as necessary.

I also question the assertion that processing score clues represents a high barrier to entry; in my experience, most people who play music are able to parse score clues comfortably and buzz on them if they know the pieces clued. If a musician were presented with a physical score, say after someone took a dictation, with a string of notes that had the right pitches and approximately the right rhythms, they would be expected to be able to at least have a good guess as to how the piece sounded. A good note clue does roughly the same thing as this. It's also worth pointing out that if players want to have more time to process score clues, or are worried about forgetting details, there's nothing stopping anyone from having paper (or even staff paper) at the game and writing down the score clue on it; I and many other players I know have done this before.

I probably have more thoughts, but on the whole, I'm a fan of note clues and find that other forms of score clues are much harder to write in an equally evocative way; I'm also a fan of score clues in general because they reward people for listening to/playing music, in the same way that descriptions of key plot elements or quotes reward people for reading books.

[Asterias Wrathbunny]

I would also posit that, despite being difficult to play, score clues actually demand a relatively surface-level engagement with subject matter

(I also just find note-listing clues aesthetically unappealing. They slow the tossup to a crawl and completely disrupt a moderator's flow!)

I think this is a gross generalization. For example, the clue "[This piece] opens with strings accompanying the notes [read slowly] “D C-sharp D E F” played on the bassoons" doesn't just reward those who have the opening to Mozart's Requiem memorized, but also alludes to the fact that the piece might be in the key of D Minor, and therefore can help players narrow down the set of possible answers. There are various ways in which a good score clue can be helpful.

[Adventure Temple Trail]

A good note transcription clue has the following properties:

• It does not reward listening to music.

I agree with the rest of these but: Huh?! I have definitely been rewarded for listening to music by "note-spelling" clues before, and I thought that was a major purpose of having them? (I will concede it is unlikely one will be rewarded in this manner without pretty good sense of pitch and/or having played some instrument before in one's life)

[dhumphreys17]

Probably one of my less popular quizbowl writing opinions is that whatever one's opinion is of note-listing score clues should also be one's opinion of computational clues. I would posit that the ability to "hear" a melody in your head based on a list of note names at game speed is similar to the ability to perform a requested calculation at game speed. I would also posit that, despite being difficult to play, score clues actually demand a relatively surface-level engagement with subject matter - akin to those annoying clues that just ask for minute, impossible-to-place details of a specific painting.

(I also just find note-listing clues aesthetically unappealing. They slow the tossup to a crawl and completely disrupt a moderator's flow!)

I 100% agree with Sarah here. Evaluating a score clue and completing a computation are eerily analogous:

1) very hard, sometimes impossible, to pyramidalize
2) tests ability to exercise skill in condensed timeframe, not depth of knowledge
3) utility of the clues is highly dependent on how the moderator reads them, more so than normal quizbowl clues

A good note transcription clue has the following properties:

• It does not reward listening to music.

I agree with the rest of these but: Huh?! I have definitely been rewarded for listening to music by "note-spelling" clues before, and I thought that was a major purpose of having them? (I will concede it is unlikely one will be rewarded in this manner without pretty good sense of pitch and/or having played some instrument before in one's life)

Anecdotally speaking, my guess is this is a perfect-pitch vs. relative-pitch thing. On the Notre Dame squad right now, we've got two or three of us that are capable of getting points in the music section of the fine arts distribution, but note-spelling score clues only help me, because having perfect pitch means I can parse them in a way my otherwise-musically-inclined teammates can't. It would very much surprise me if someone has been helped by a note-spelling clue because they had read the underlying piece of music without having internalized what the notes on the page actually sound like, but power to anyone using note-spelling clues like that, pardon the pun.

[AKKOLADE]

If note spelling clues require a sense of pitch to buzz on them - which seems fairly innate - should that be knowledge that is tested in a quiz bowl question, particularly a tossup?

[Halinaxus]

3) utility of the clues is highly dependent on how the moderator reads them, more so than normal quizbowl clues

At my Winter site last weekend, one moderator was "singing" the score clues as he read them, something I don't think I've ever had a mod do before. I know basically zero afa so this didn't change my experience at all, but I imagine the players who did know music had an easier time parsing and buzzing on the questions with score clues in this particular room because of it.

Obviously, not all moderators are going to be able to "sing" the score clues like that (I certainly wouldn't be able to), which creates an inherent (and I presume undesirable) difference in difficulty between different rooms at a tournament.

Like I said, I'm not a music player, so I don't have an opinion on score clues in general, but this particular experience did strike me as somewhat problematic.

[Adventure Temple Trail]

At my Winter site last weekend, one moderator was "singing" the score clues as he read them, something I don't think I've ever had a mod do before...Obviously, not all moderators are going to be able to "sing" the this particular experience did strike me as somewhat problematic.

Agreed. Don't do this.

[Subotai the Valiant, Final Dog of War]

If note spelling clues require a sense of pitch to buzz on them - which seems fairly innate - should that be knowledge that is tested in a quiz bowl question, particularly a tossup?

Anecdotally speaking, my guess is this is a perfect-pitch vs. relative-pitch thing. On the Notre Dame squad right now, we've got two or three of us that are capable of getting points in the music section of the fine arts distribution, but note-spelling score clues only help me, because having perfect pitch means I can parse them in a way my otherwise-musically-inclined teammates can't.

Two notes:
1. Sight-singing, or being able to reproduce pitches in the proper rhythm given a starting note, is a skill that musicians practice and learn to be able to do. It is perhaps easier for some people to learn than others, particularly people with perfect pitch, but it's not something that is innately impossible (or something people innately are able to do). In fact, it's something that the College Board expects that people can learn to a passable level in one year, as it's required by all takers of the AP Music Theory exam! I certainly don't think that AP Music Theory is perfect, and I question that everything on it can actually be learned in a year, but I think it's fair to say that the testwriters wouldn't put anything on it that is completely impossible to learn within a year or two to a passable level.

2. If this is a problem people are seeing consistently, we could try to start writing more score clues in terms of scale degrees, or intervals, or chord types. Then everyone's on the same playing field. This also has the effect (which can be sometimes desired, other times not) of not giving away the key of the piece in question or privileging the appearence of a repeated theme in one key over another. I am a big fan of doing this in certain circumstances and think it would be a positive change for certain types of clues, especially ones where harmony matters more and the importance of the clue is the harmonic progression rather than the exact chords.

Also, note-spelling clues/score clues should reward people for having internalized the music? And is technical knowledge of the field not a requirement to fully understand the substance of a clue in many other categories? (Many science questions, for example, are utterly incomprehensible within their first few lines without a year, or several years, worth of college math or science courses, plus years of high school math and science building up to those)

[warum]

2. If this is a problem people are seeing consistently, we could try to start writing more score clues in terms of scale degrees, or intervals, or chord types. Then everyone's on the same playing field.

As someone who has decent relative pitch but not absolute pitch, this would not help me. When I hear a "note clue" spelled out, I usually imagine the melody without thinking of a particular key at all. Also, saying a series of intervals is fine for 3 or 4 notes but gets cumbersome for anything longer.

One thing that sometimes prevents me from buzzing on a "note clue" is lack of information about where the notes being spelled begin, relative to the meter of the piece. There have been tossups at multiple tournaments in the past year where I imagined a "note clue" starting on the first beat of a measure and could not recognize it, only to find after the tournament that the clue actually begins on a pickup. This issue is just one aspect of why giving context for a note clue is so important.

[modernhemalurgist]

I guess I’ll weigh in on the pro-note spelling side. I don’t really have a cohesive argument, but here is a loose assemblage of rambling thoughts.

• I don’t quite buy the argument that score clues are comparable to computation clues. The latter are bad because they require doing nontrivial mental work in your mind, and they ultimately just test your ability to do this at speed. Converting score clues does, I admit, take a bit more time than just recalling stock facts, but nowhere near as much as computation, especially when read slowly (more on this below). Certainly clues in many science questions, as others have pointed out, require at least as much effort of “translation.” Really as long as you can process what each note is, or ideally write them down, I think good score clues can be pretty instantaneously buzzable.
• I have an alternative comparison: score clues in music tossups are analogous to plot clues in lit tossups. Both test “surface level” understanding of a work, i.e., do you know the most famous/memorable/unique things that happen in this work? Certainly they shouldn’t be the only thing asked about (though I have often facetiously argued as much), but they are often the most buzzable things other than directly naming composers or works. Clues from formal structure, history, analysis, etc. of a piece, just like clues from literary criticism or history, should be used sparingly to test the deepest level of knowledge, and can rarely serve as the backbone of a tossup.
• Score clues are easier than I think many people give them credit for. I’ve noticed this most prominently when reading trash questions with score clues, where my generally non-musical teammates can get very good buzzes from a string of pitches. The most important factor is not how well-developed your sense of pitch is (although this helps), but how well you know the works being clued. I say this as someone who has, for a musically-inclined person, a pretty terrible sense of relative pitch. Most of my buzzes on score clues come either from having seen the score, or from a combination of the general contour of the clue and knowing the key. The latter is significant musical knowledge (which, I think, we should be testing), but the former is the more important part, and in many good clues is very buzzable.
• In fact, I usually find that note spelling clues are, to reiterate Chris’s point, generally the easiest kind of score clues. Sometimes specific orchestration clues are identifiable, especially at lower difficulties (and especially memorable things like offstage instruments, unusual/archaic instrumentation, etc.), but for most tossups descriptions like “string arpeggios underpin a soft trumpet solo” are just terribly difficult to pin down. Musical texture is rarely so unique and evocative as to be able to narrow in on one piece from a description of a single moment.
• I think many have rightfully pointed out that moderators are terribly inconsistent about reading strings of pitches. I’ve often had a moderator blaze through these clues at full speed even when there is a note in the packet to slow down, or worse, marked pauses to indicate rhythm. Beyond this there can be issues with enunciation of pitches, and apparently with moderators singing the clues (I’ve never seen this happen but am aghast at the idea). I’m not sure what can be done about this - make it clear in the packet, or at tournaments, that packet directions should be taken seriously? In any case it doesn’t seem like an argument against note spelling clues, it seems like an argument for better moderator training.
• I will say that I don’t quite agree with Chris’s point on length. Unfortunately, long score clues definitely do drag on and bog down the pace of a match, and so it’s important to choose note strings that are recognizable from only a few pitches (I’d say around 6 at maximum). This is a rather artificial limitation that I don’t love, but I do think it’s important for preserving pace and feel. That being said, longer pitch sequences could serve as great clues in bonuses, where pace matters a bit less. I do think this doesn’t happen enough - bonuses are a great place to isolate knowledge of a piece, where processing time doesn’t matter quite as much.

[Gene Harrogate]

I have an alternative comparison: score clues in music tossups are analogous to plot clues in lit tossups. Both test “surface level” understanding of a work, i.e., do you know the most famous/memorable/unique things that happen in this work? Certainly they shouldn’t be the only thing asked about (though I have often facetiously argued as much), but they are often the most buzzable things other than directly naming composers or works. Clues from formal structure, history, analysis, etc. of a piece, just like clues from literary criticism or history, should be used sparingly to test the deepest level of knowledge, and can rarely serve as the backbone of a tossup.

I don't know about this. I think everyone agrees that recognizing note-spelling clues requires some degree of technical training besides just having listened to a piece of music. The plot of a novel is available to any literate person (i.e., any person who plays quizbowl).

I'll throw in a bit of anecdotal evidence in the other direction, which is that I personally don't know many people (including those with some musical training) who are able to buzz on these clues outside of having seen them before in quizbowl packets. In general, I think people who are deeply ensconced in music sometimes underestimate how much effort it takes for dummies like me to aquire the skills they think of as basic.

[Cheynem]

I am literally tone deaf, so perhaps not a good example, but I listen to a fair share of music, classical and popular, and have no idea what any note clue means. That's not an excuse obviously to end such clues, but I don't think it's as immediately easy to pick up on as any other clue.

[AKKOLADE]

Two notes:
1. Sight-singing, or being able to reproduce pitches in the proper rhythm given a starting note, is a skill that musicians practice and learn to be able to do. It is perhaps easier for some people to learn than others, particularly people with perfect pitch, but it's not something that is innately impossible (or something people innately are able to do). In fact, it's something that the College Board expects that people can learn to a passable level in one year, as it's required by all takers of the AP Music Theory exam! I certainly don't think that AP Music Theory is perfect, and I question that everything on it can actually be learned in a year, but I think it's fair to say that the testwriters wouldn't put anything on it that is completely impossible to learn within a year or two to a passable level.

I didn't realize that - I have no formal music education beyond a 100 level appreciation course. Thanks for the clarification.

[Mike Bentley]

Score clues are easier than I think many people give them credit for. I’ve noticed this most prominently when reading trash questions with score clues, where my generally non-musical teammates can get very good buzzes from a string of pitches.

I would be shocked if at a typical trash tournament (which already has a very selected for quizbowl audience) that note-spelling of key riffs from even the most iconic songs are getting anywhere near the buzzes of other types of clues.

[Subotai the Valiant, Final Dog of War]

Two notes:
1. Sight-singing, or being able to reproduce pitches in the proper rhythm given a starting note, is a skill that musicians practice and learn to be able to do. It is perhaps easier for some people to learn than others, particularly people with perfect pitch, but it's not something that is innately impossible (or something people innately are able to do). In fact, it's something that the College Board expects that people can learn to a passable level in one year, as it's required by all takers of the AP Music Theory exam! I certainly don't think that AP Music Theory is perfect, and I question that everything on it can actually be learned in a year, but I think it's fair to say that the testwriters wouldn't put anything on it that is completely impossible to learn within a year or two to a passable level.

I didn't realize that - I have no formal music education beyond a 100 level appreciation course. Thanks for the clarification.

No worries! And this wasn't meant to be targeted at you or anyone else; I've heard similar sentiments from a number of players in the past so wanted to just clear the air for anyone reading that parsing notes/sheet music is something that takes time for everyone but is something that can come with time for most people.

[alexdz]

I've engaged directly in the performance of many, many pieces of music throughout high school and college. It might seem as though score clues would be a big help to me. Yet, I can almost never make sense of them because of the reverse of Christopher's comment:

... if you're confused by transposing instruments, it can help to look up the same passage in a piano transcription so as not to get confused by which key you should be writing them in.

Despite playing my instrument proficiently for 11 years, I struggle with score clues because of the specific instrument I play. Since clarinet music is transposed, any sequence of notes I am used to seeing on a page of music I've played won't have been written in concert pitch.

I will admit, I'm not sure how to think about that in terms of coming down firmly for or against score clues. That said, if as has been stated upthread that we want to reward deep engagement with music (i.e., having performed a piece), writing score clues in concert pitch does seem to disadvantage those who might have learned a piece on a transposing instrument.

[AKKOLADE]

Two notes:
1. Sight-singing, or being able to reproduce pitches in the proper rhythm given a starting note, is a skill that musicians practice and learn to be able to do. It is perhaps easier for some people to learn than others, particularly people with perfect pitch, but it's not something that is innately impossible (or something people innately are able to do). In fact, it's something that the College Board expects that people can learn to a passable level in one year, as it's required by all takers of the AP Music Theory exam! I certainly don't think that AP Music Theory is perfect, and I question that everything on it can actually be learned in a year, but I think it's fair to say that the testwriters wouldn't put anything on it that is completely impossible to learn within a year or two to a passable level.

I didn't realize that - I have no formal music education beyond a 100 level appreciation course. Thanks for the clarification.

No worries! And this wasn't meant to be targeted at you or anyone else; I've heard similar sentiments from a number of players in the past so wanted to just clear the air for anyone reading that parsing notes/sheet music is something that takes time for everyone but is something that can come with time for most people.

No offense taken - I literally asked the question, after all!

[Eddie]

A good note transcription clue has the following properties:

• It does not reward listening to music.

I agree with the rest of these but: Huh?! I have definitely been rewarded for listening to music by "note-spelling" clues before, and I thought that was a major purpose of having them? (I will concede it is unlikely one will be rewarded in this manner without pretty good sense of pitch and/or having played some instrument before in one's life)

Whoops, you are completely right, and I recant this— I had been wavering between making a list of properties of a "good note transcription clue" and an "average note transcription clue", and I accidentally left "does not reward listening to music" under the "good" category when I submitted my post.

A more accurate statement of my position would be that a good note spelling clue may reward listening to music. In the line of Matt's parenthesised concession and Mike's post about tonedeafness, I think that "listening to music" is not the primary skill or learning method that is rewarded by note spelling clues, but rather the related skills of (a) quickly parsing a string of notes, (b) mentally converting the string of notes into a musical passage with reasonable precision, and (c) finding a binary match between the musical passage and the desired answer. To this end, I would like to again place myself in the side of people who think that note spelling clues are comparable to computation clues, and rebut the below quote.

• I don’t quite buy the argument that score clues are comparable to computation clues. The latter are bad because they require doing nontrivial mental work in your mind, and they ultimately just test your ability to do this at speed. Converting score clues does, I admit, take a bit more time than just recalling stock facts, but nowhere near as much as computation, especially when read slowly (more on this below). Certainly clues in many science questions, as others have pointed out, require at least as much effort of “translation.” Really as long as you can process what each note is, or ideally write them down, I think good score clues can be pretty instantaneously buzzable.

Though I agree that there is a difference between the specific skills of "converting a string of notes into a musical passage" and "doing mathematical computation", I think that there is a more abstract level at which these skills both require the prospective buzzer to carry out rapid symbolic manipulation. Moreover, I think that what counts as "trivial" or "nontrivial" is highly subjective and difficult to quantify. If there is at least a sizeable minority of people for whom buzzing on a note spelling clue is deemed to behoove "nontrivial mental work", I think that this is the skill floor at which we need to judge note spelling clues. In other words, the bar for a good note spelling clue should be "whether an average musically-literate quizbowler can buzz", and not "whether a highly-practised music specialist quizbowler can buzz".

[34 + P.J. Dozier]

I would probably tend to agree with the anti-score clue arguments, which is ironic considering I've written and edited a lot of music questions over many years, and most of those questions probably contain at least one score clue. As someone with perfect pitch as well as sight-reading and sight-singing ability (as well as a lot more experience with music in general than the average quizbowler), I have almost never found score clues helpful, and when I have buzzed on score clues in the past, it's because I was buzzing on a quizbowl clue that had come up before.

That being said, I do think that score clues can be done well (and they have been done well by certain writers, who I look up to a lot in this regard), but most quizbowl writers are unable to write score clues well because they don't have the requisite understanding of music to do so. I don't want to sound gatekeep-y when I say this, but I feel as though most quizbowl music writers are people who think that they really understand music just because they played in orchestra in high school and carded a bunch of score clues. Those people do understand music more than the average person, but they lack an innate intuition for what is important and helpful to musicians' ears, because their intuition for what is important and helpful was molded by quizbowl questions, and not by a strong engagement with all kinds of music throughout their whole lives. In a way, it's like when you can tell if a science question was written by someone who actually is a real scientist with lab experience and such, versus if a science question was written by someone who became a really good science player through quizbowl learning. I feel as though that is why a lot of science tossups drop trivial doubly-eponymous equation names, and that is also why a lot of music tossups drop disjointed and useless score clues.

My personal philosophy on score clues being done well, I think, can be most appropriately analogized to most people's philosophy on visual art clues being done well. When one looks at a painting, one can generally point out some of the most notable aspects of the painting –– its central figure(s), color palettes, important background details, artist signatures, etc. Besides those notable aspects, there are hundreds –– perhaps thousands or more –– of aspects that are relatively less notable, right down to the minute details of a single brushstroke. When you zoom in on that brushstroke, of course those minute details seem notable –– much like when a music question writer goes hunting for a clue from a symphony, presses "play" on one of the movements, and then listens intently for some melody, any melody, to latch onto and write a score clue about. But just because you're cluing a melody doesn't mean that you're cluing something that is notable and important –– there are hundreds of melodies in one symphony, and only a handful of them are as notable and important as a central figure of a painting. People who have engaged with and understand music deeply will have the perspective and intuition to listen to something and identify it as an important line in the grand scheme of the work; people who possess a relatively surface-level proficiency with music do not have that same acuity.

[theMoMA]

Score clues are easier than I think many people give them credit for. I’ve noticed this most prominently when reading trash questions with score clues, where my generally non-musical teammates can get very good buzzes from a string of pitches.

I would be shocked if at a typical trash tournament (which already has a very selected for quizbowl audience) that note-spelling of key riffs from even the most iconic songs are getting anywhere near the buzzes of other types of clues.

I can only assume this is a veiled slight on the notes clues from the Power Rangers flute dagger in the CO Trash tossup on Zords, which I can safely report generated at least one buzz.

[Asterias Wrathbunny]

I want to add that in my experience, most note strings are accessible, even at the highest levels of play. The Wikipedia articles for a great many pieces contain score excerpts of the significant musical elements (e.g. themes from a symphony), and you just need to be able to read music to be able to learn them. I can’t really think of an instance where a score clue unfairly benefitted someone with perfect pitch by including a string of notes that was not emphasized in a study source like Wikipedia or Cambridge music handbook.

In addition, I’ve buzzed on various score clues, note strings or otherwise, throughout my time playing quizbowl, and I can safely say that most of them come from having viewed parts of the score or watched a video of someone playing the piece, not from perfect pitch.

[ThisIsMyUsername]

My, what a busy thread! And one I’m glad to see attracting so many posts.

As those of you who are older may know, I spent the first half of my quizbowl career vigorously lobbying for the inclusion of “score clues” (broadly defined) and for the reform of how we wrote them, to make them describe less obscure features and describe them more precisely. Things have come a very long way since I wrote my music-question-writing guide in 2014. The sample howlers I included from that year’s Nats are of a kind that one virtually never sees in tournaments of recent years. So, whatever misgivings I may have about current practices are easily outweighed by my happiness at how generally healthy score-clue-writing is, these days.

With that said, I have to agree with what Matt Jackson said in his first post:

In recent years, it seems as though the term "score clue" has undergone semantic narrowing to the point that it almost always means "transcribed string of notes", and that kind of clue ("in this piece, the trumpet melody goes D-E-F-F-etc.") has also become significantly more common. I have sometimes been helped by these clues, and I believe there should be some sometimes. But I find it unfortunate that the "note spelling" subtype of score clue seems to be crowding out other kinds of clue in auditory fine arts questions.

I quote his actual words, because it’s interesting to see that the first lengthy response in this thread (Chris Sims’s) ends with a defense of score clues writ large. This helps to prove Matt Jackson’s point. Only in an environment in which the term “score clue” has undergone precisely the sort of “semantic narrowing” he is talking about could one think that a call for fewer “note-spelling” clues is some kind of attack on the institution of score clues in general! (We see this also in Wonyoung and Alex’s replies, which also address score clues in general, as if this were the same subject.)

It seems to me that there are at least three strands of discourse running through this thread:

- Empirical claims about whether note-spelling clues are more buzzable or less buzzable than other types of music clues
- Value claims about whether the skills and knowledge involved in buzzing on note-spelling clues are the kinds quizbowl should focus on rewarding
- Practical suggestions on how to modify existing note-spelling clue practices in service of one of the above two domains—that is, either to make them more buzzable or to make them more rewarding of some approved form of knowledge

The empirical claims are basically all anecdotal. (“I can / I can’t / most people can / most people can’t buzz on note-spelling clues!”) In response to them, I’d like to see some data from tournaments that kept buzzpoint data. Are note-spelling clues generally more buzzable or less buzzable than non-note-spelling clues in the same positions in tossups? Are longer note-spelling clues more helpful or less helpful than shorter note-spelling clues? It might be that the sample of note-spelling buzzes we can find are too small to be used as good data. But if so, that would prove the point of those who think that note-spelling clues are not super buzzable.

Some of the value claims have centered on Sarah’s analogy between note-spelling and math computation. It’s strange to me that people have taken this bait. It doesn’t matter if the fundamental skills might be similar in some way; the latter is the exercise of a skill for exercise’s sake, whereas the former is the exercise of a skill in order to process a description of the core features of an artwork. Whatever “calculating” might be happening is a means of recognizing and engaging with the aesthetic content of the musical work being clued. This difference alone makes the analogy completely incapable of proving anything about the value / lack of value of note-spelling clues.

When Sarah says something like “score clues actually demand a relatively surface-level engagement with subject matter - akin to those annoying clues that just ask for minute, impossible-to-place details of a specific painting,” it should be even more obvious that we’re not dealing with the reality of how music questions are currently written. Most note-spelling clues in the past five or so years are about literally the most immediately recognizable melodies in a piece; they thus bear no resemblance whatsoever to “minute, impossible-to-place details” in a painting. If they have any correlates in visual art, those would be the central figures of a painting, not background details. Likewise, when Devin criticizes note-spelling clues because they test “ability to exercise skill in condensed timeframe, not depth of knowledge,” he’s making a similar error. These kinds of clues are not inherently about something surface or deep. Note-spelling is a mode of description. The thing being described can be ultra famous or maddeningly obscure.

Continuing on the subject of value, it’s worth noting that Matt Jackson never argued that note-spelling clues aren’t about important parts of pieces. He argued that they are crowding out the other forms of musical knowledge worth rewarding, some of which were enumerated by Ophir in the list that Matt provided. It’s sad to me that there hasn’t been more acknowledgment of the necessity of pursuing those other subjects.

However, it is the practical matters that interest me most. Here's what I regard as the important point: Not all themes translate equally well into score clues, and not all themes that might translate equally well into score clues translate equally well into pitch-based score clues. One could say more generally: themes that would be equally buzzable if played/heard in a music-listening tournament will not necessarily be equally buzzable when described as some sort of score clue, even if the score clue is ideally formulated.

This is crucial because it is where the analogy between something like literature-writing and music-writing breaks down. For example, for most medium-length poems, if I choose the right lines and quote them in the right order, I can automatically produce a playable tossup. I cannot do this for a piece of music. I cannot take theme after theme—even exclusively the most famous themes from a piece—spell them as pitches, and produce a series of viable clues. Each theme is partly constituted of several non-pitch elements, including rhythm, instrumentation, dynamics, tempo, etc. For some of these themes, the inclusion of the right combination of these elements is necessary for a clue to work.

Thus, while I’m alarmed by the over-preponderance of “note-spelling” clues (or of clues that use non-pitch information primarily for context, not as a fundamental part of the clue), this is not primarily an ideological objection. It seems pretty clear to me that the reason these clues have become so popular is that writers are not giving enough thought to selecting the right elements to effectively describe any given musical moment. (Rattling off a bunch of pitches is possibly the most brainless way to write a music tossup.) I am not convinced that melodies should be the default form of score clue (although I regard that view as defensible); but even if they are, bald “note-spelling” should be a small piece of what this involves, not the entirety or near-entirety.

For example, I found what I believe to be the Scheherazade tossup that Chris was referring to. Here it is:

The theme of this work’s second movement first appears in a grace-note-littered bassoon solo of a descending sequence of rapid major seconds beginning F-sharp, E, F-sharp, accompanied only by muted double basses. The opening of this work’s third movement features 26- and 32- note tuplets in the clarinet and flute, and a snare drum accompanies a more staccato variation of its main theme in a triple-p clarinet solo beginning C, D, long E, D-sharp, E. The descending whole (*) tone scale E, D, C, B-flat is emphasised in a fortissimo pesante [pe-zan-tay] theme which contrasts, and eventually reconciles, with a recurring violin recitative of semiquaver triplets representing this work’s title figure. With movements like “The Kalendar Prince” and “The Sea and Sinbad's Ship”, for 10 points, name this Rimsky-Korsakov suite named for the narrator of A Thousand and One Nights.
ANSWER: Scheherazade [or Shekherazáda]

It’s striking to me that the first two clues are much better than the next two. Even though the description “descending sequence of rapid major seconds” is inaccurate for multiple reasons, the combination of grace notes and bassoon accompanied only by only muted double basses (a good instrumentation clue, because this is such an unusual combination) would allow me to go in on “F-sharp, E, F-sharp.” For the second clue, 26- and 32- note tuplets in the clarinet and flute are more distinctive than virtually any pitch string could be. (Although, saying each of these tuplets forms a rising and falling arc would seal the deal for me.)

However, at least for me, the third clue is hampered by the fact that it relies too heavily on the pitch (and has accidentally provided the notated pitch in the B-flat clarinet part, rather than concert pitch). Chris has suggested lengthening the string of pitches. But I’d say that this is the perfect example of a clue that needed a rhythm rather than more notes. The most notable thing about this movement is the rhythm of the two themes (indeed that’s what most strongly links them, to begin with). Even if you gave me only “In 6/8, two pickup eighths, then dotted-eighth, sixteenth, two tied eighths” that would be significantly more buzzable than this string of pitches, and that rhythm plus some combination of clarinet, snare drum, and dynamics (the other information given in the actual tossup) is more than sufficient. (The reason the fourth clue is not ideal is that it provides the notes of Sharyar’s theme in scale order, rather than in the order that they appear in the actual theme.)

Natan has referred to this problem of missing rhythm as well. I would suggest that writers apply the following test at minimum: If you’re going to give a string of pitches without rhythm, force yourself to hum aloud (or play) the string of pitches at an even rhythm, with the metric stress on the first one. How recognizable is this melody to you? If you have any doubts about this, add other elements! Even better, don’t make starting from pitch alone your default! Consider writing out, for yourself, all of the most helpful elements and subtract until the clue is short enough.

People have also raised the problem of recognizing whether an interval is supposed to be ascending or descending. I failed to buzz once on one of my favorite motives in classical music, because I was mentally seeing/hearing it in the wrong direction! I thought JORDU had the most brilliant solution to this: it declared at the tournament’s outset that all note-strings assume movement to the nearest note of the same letter, unless stated otherwise. Thus, C-E would be ascending by default, while C-G would be descending by default. I considered doing this for ACF Nationals 2022, but ultimately did not, because I wasn’t sure if it was within ACF rules to make this sort of announcement in Round One of a national tournament. I think this is something that tournaments should strongly consider doing in future. If they don’t, it is absolutely incumbent on the question to specify the direction in cases where a mistake is likely and impactful.

Additional thoughts from me later, perhaps.

[Heiliger Dankgesang]

I don’t have much to add to what’s already been said, since I agree with many of the points that have already been outlined in this thread. But there is something in here that I’d like to reiterate.

I am a fan of combining notes and other information as best as possible, since more useful information makes it more likely that any given person can buzz on the clue with the same amount of existing knowledge; in that sense, I'm not actually a fan of pure note clues. But I also don't think there are that many questions where pure strings of notes and rhythms are being listed without any additional context.

I’ve always gravitated toward classical music questions precisely because I am a classical musician. And I have always found score clues fun specifically because it is easier for me to buzz on them due to having perfect pitch (it’s helped me way more often than not). But while score clues do reward listening to music or studying a score, that is all that they often accomplish. Score clues often aren't deep cuts about a piece as much as that they identify how a piece sounds, essentially like a game of “name that tune.” Without getting into details, I have since been annoyed by questions whose sole premise is note- or rhythm-spelling because they are often less substantive than questions that don’t contain score clues at all. (I have talked to other players who share the same sentiment.)

I definitely don't think that standalone score clues are bad; I still enjoy them because they are important in their own right— they are the purest way of describing a piece of music, and they are particularly effective when they have evocative descriptions of how notes, rhythms, and tempi, etc. are combined. (On that note, they indeed aren't surface-level even if they may appear so.) However, on that note [pun intended], a combination of notes/rhythms/tempi/etc. does not need to be the most important part of a score clue. Score clues are much more meaningful and interesting if they are accompanied by some sort of context. It’s also worth noting that for score clues that don’t contain sequences of spelled-out notes (which I agree are less reliable to confidently buzz on when isolated), writers and editors should make an even more concerted effort to make those clues accessible; context can help make those clues concrete. And while it is important to give as many notes as possible for a player to be able to buzz confidently (such as with Chris’s Scheherazade comment), I prefer to not include so many notes in a given score clue because then they will occupy too much space in a tossup at the expense of other information in the clue that may be more helpful. (Also, as others have duly noted, one can't underestimate the importance of including rhythmic information and relative information about pitches besides note names.)

[Note that the examples below and the comments that follow solely illustrate some of my goals for context in score clues and may not reflect what players thought while hearing each question. These examples are also neither ideal nor complete, and while I was looking for them in 2021 ARCADIA, I noticed several instances in my editing of score clues where I realize that I definitely could have done better. There are examples from this year that I would have rather included, but I suppose you will just need to play 2022 ARCADIA to hear them *wink*.]

In the finale of a piece inspired by this person, the pizzicato notes “E-flat, high B-flat, low B-flat, E-flat” start a theme and variations from The Creatures of Prometheus. (Packet 12, TU 8; Answer: Napoleon)

In this G-sharp minor piece, the right hand leaps between the melody, played with thumb, and a high D-sharp to depict the title ringing object. (Packet 3, Bonus 15M; Answer: La Campanella)

In the Napoleon tossup, people may be thinking of the main theme from Movement 4 of the “Eroica” Symphony after hearing its first four notes, but that thought may be solidified upon hearing immediately afterward that the spelled notes are based on something from the Creatures of Prometheus. The “La Campanella” bonus part balances performance technique, notes, and the purpose that those both serve: ringing a metaphorical bell.

In the end, it’s best to ask why a particular score clue is an important and defining feature of a piece. Is it a unifying motif or cipher? Is it intended to evoke something beyond the piece itself? What inspired it? Is there a particular directive that singles out those notes and underscores their role in a piece? Notes should not be spelled out just for the sake of cluing them. A score clue should by all means reward people who have heard, performed, or score-studied a piece; however, at the same time, it should be a vehicle that helps reveal something about that piece that is a bigger think– something more accessible that more players are likely to know more confidently.

[EDIT: I added/reconsidered some points after reflecting on John's post to take it into account; I did not see that post appear while I was busy writing my own post.]

[setophaga]

Bit late here, but making one of my very infrequent visits to the forums and happened to see this.

As others have said, there's a big difference between "note-spelling" and score-clues - in fact, you could really call anything relating to the performance aspect of a piece, outside of its musicological or popular interest, a score clue! I highly encourage music writers to use instrumentation, form, textural, and metrical clues in addition to melodic spellings.

I tend to agree that note strings should be short, iconic, and infrequently placed. They are generally hard to parse even with absolute pitch, and I say this as both a professional performer and avid listener. As others have said, they can be highly moderator-dependent; a poor moderator is going to have a lot of trouble fairly delivering the clue to a comprehensible degree.

Again, this should not be an excuse to avoid or reduce performance clues; performance and listening knowledge (in addition to purely academic knowledge of theory, history, and aesthetics) should be rewarded when we write music questions.