How many tunes are there?

How many tunes are there?

Maybe the question has come up before, but I don’t recall seeing it over the past six years, and a search of the discussions for "how many tunes" didn’t find anything on the subject.

Okay, I know we can’t answer the question. But I am interested in what people might give as a rough estimate. Personally, I think its thousands. That’s pretty rough because it runs from two thousand to nine hundred and ninety-nine thousand, and there is a significant difference between the two. Then again, I’m not very knowledgeable on the subject so I can’t afford to be much more precise.

By tunes, I mean traditional Irish music tunes (including the ones whose authors we know) that might reasonably be played at a session and recognised by at least one other person besides the person who starts it. Although it might be necessary to tweak that definition a little given the promiscuous transfusions between the Irish, Scottish, Shetland and even Breton and Asturian/Galician traditions, not to mention a few others.

This is not research for an academic paper. I am just killing some time this afternoon because I haven’t got any work on my desk right now and its a thought that occurs to me from time to time, often in conjuction with the fact that six years after starting both the fiddle and traditional Irish music from scratch, I am now approaching around two hundred tunes that I would not hesitate to join in with at a session, although a lot fewer that I would be willing to start myself. Sometimes it seems that some fiendishly clever demon is sitting somewhere churning these things out for the sole purpose of making sure that I will always, no matter how long I live, have a lot more tunes to learn.

So, any ideas?

Re: How many tunes are there?

I haven’t the foggiest!
But this just reminded me of a time a saw this guy with a t-shirt I found quite amusing. On the front it had a wee mouse playing a fiddle by candle light and on the back it said "So many tunes… so little time." I just thought… oh my gosh… so true!!

Posted by .

Re: How many tunes are there?

I just ran a search on this website, leaving all the fields blank.

861 pages of 10 tunes each gives 8,610 tunes.

That is a poor estimate, as it doesn’t include all the tunes that are played and are not on this website.

"… making sure that I will always, no matter how long I live, have a lot more tunes to learn…"

…and that, is something to be very, VERY thankful for, as far as I’m concerned! πŸ˜‰

Re: How many tunes are there?

here’s a thought….

mathematically, there must be a finite number of ways to arrange the notes of two 8 bar phrases

it would be a large number, but it exists

I call this "Nate’s Theorem", but I’m still working on the proof

but what Nate’s Theorem gives us is the upper limit on how many tunes could exist.

Nate’s Second Theorem is that the actual number of tunes played at session in any given era or in any generation is less than the number of possible note conbinations given in Nate’s First Theorem.

Re: How many tunes are there?

My abctunesearch.com site currently has 12860 rows in the tune database. A good number of those are duplicates (different settings of the same tune, from different sources, etc…)

Now, this includes an awful lot of tunes that wouldn’t fit within your criteria of tunes that might be "reasonably played at a session". Although, every session is different, and pretty much every tune has probably been played at a session at one time or another. My session sometimes plays tunes written by members of the session. Other people will know the tunes, and play along, but no other sessions would know those tunes…

And it is not unheard of for someone to be looking for a tune that isn’t in my database. In fact, it happens more often than I ever would have thought.

So if I had to estimate, I would say… 12. There’s really only 12 tunes, and lots of different settings πŸ˜‰

Re: How many tunes are there?

Krick’s theorum:

number of tunes= number of tune names/ 10

Re: How many tunes are there?

8

Posted by .

Re: How many tunes are there?

"A serious Irish musician lives in a state of permanent low-level anxiety over how many tunes she doesn’t yet know."

- ‘Field Guide to the Irish Music Session’, Barry Foyle

Re: How many tunes are there?

Didn’t someone say once that there are only seven or eight stories - and that all stories are variants of those? Is that the same with tunes?

Re: How many tunes are there?

Re "Nate’s Theorem"

If you allow for the fact that there will be a certain amount of repetition in a 16 bar tune, and assume a fair distribution of crotchets and quavers, and then just calculate on the basis of each note being Same, Higher or Lower compared with the previous one, you quickly run into the billions of potential tunes.

It all becomes a bit of a waste of time!

Re: How many tunes are there?

Yeah, I’m afraid that Krick’s Theorem is probably pretty accurate, but fortunately for us, the tune names are irrelevant anyway. But hold on just a minute, does that give us a 0 in the dividend of the equation? Uh-oh.

And yeah, Reverend, Mark and maxF, I am reminded of this thing that I read once about all the electrons in the universe being exactly identical, leading some waggish physicists to suggest that there is in fact only one electron, it just gets around.

Re: How many tunes are there?

Reverend…I see that I was premature in naming the number of possible note combinations "Nate’s Theorem". As you have done the proof and calculated the number, it should be your right to name the theorem.

Re: How many tunes are there?

my exwife thinks there is only one.

Re: How many tunes are there?

Wasn’t it Abe Lincoln who said he only knew two tunes: "One is Dixie and the other one isn’t."

Posted by .

Re: How many tunes are there?

that was General US Grant

Re: How many tunes are there?

>> it should be your right to name the theorem

OK, I officially dub it "Nate’s Theorem" πŸ˜‰

Re: How many tunes are there?

most people reckon there’s only one irish diddley tune. I once told someone I just played the same tune over and over again all night - and they believed me

Posted by .

Re: How many tunes are there?

I don’t remember where I read this but the estimate was between 10 000 and 15 000 tunes.

Posted by .

Re: How many tunes are there?

Nate’s or Reverend’s theorem might give a finite number of possible permutations, but we all know that only a tiny fraction of those (albeit still a vast number) would actually sound anything like Irish Traditional Music - and the parameters that make a tune sound ‘trad’ are far too multifarious, vague and subjective to work into the theorem, I suspect. Somebody with a big enough brain might be able to design some software that, fed a few thousand tunes, could work out some kind of universal pattern or group of patterns and thus, the number of possible tunes that conform to those patterns. But, even then, there will be rogue tunes that somehow get away with not fitting the mould.

I think it was Breandan Breatnach who reckoned there to be round about 8000 tunes in circulation among Irish traditional musicians. But that was 30 or more years ago. I suspect that nowadays the collective repertoire assimilates new tunes at a much faster rate, with recording artists composing hundreds of new tunes and adopting and adapting tunes from other traditions, and would thus be somewhat larger. And, with the benefit of archive recordings, older tunes that have fallen out of circulation can be rediscovered and re-introduced at any time. But I have a hunch that, if we set the statisticians to work on this, they would find that there is some kind of critical number of tunes, which remains more-or-less constant, regardless of how many traditional musicians there are in existence (There may be more people playing this music now than ever before, but most of them play together in large numbers and have broadly overlapping repertoires, whilst in the past, musicians more often played alone or in small numbers and had their own personal repertoires, with only a little overlap), or how many new tunes are composed (Only a small proportion of those stay in circulation for long).

Just a few idle thoughts. I ain’t no scientist.

Re: How many tunes are there?

"861 pages of 10 tunes each gives 8,610 tunes."

That figure is misleading, as it includes all the alternate titles (of which there might be 5 or 10 for some tunes), as well as quite a few duplicated submissions that have not been deleted.

Besides, there are only 8 tunes on the final page, so it’s 8608.

Re: How many tunes are there?

Not to mention that some fraction, not insignificant are going to consist of newly-composed and never-played tunes… and that fraction is increasing, it seems, daily. I see someone’s posted a tune with an extra half-part - bet that one gets played plenty…

Re: How many tunes are there?

Warren Spahn said that a pitcher needs only two pitches: one the hitter is looking for and another one to cross him up.

If a similar rationale were applied to music making, then there might only be two tunes: the one you’re playing and the one you can’t remember.

Of course, there’s always the next one…

Re: How many tunes are there?

Based on the starting definition, I’d say the number is south of a thousand, increasing or decreasing as we twiddle some of the parameters. Now, of course there are many more than a thousand tunes, but tunes that are part of the repertoire of at least one session, today? Much smaller number.

To begin with, I want to adjust the definition a little. As it stands, the question is really, "how many tunes are known by at least two players today?" That seems a broader question than the one you’re asking, which is, "how many tunes are being played today?"
So rather than asking in terms of the hypothetical, "how many tunes T are there such that I could walk into some session somewhere and play T, and have someone else know it", I want to cast it in terms of "how many tunes were played in sessions in the last month?" (or, if you’d rather , in the last year, I think the numbers will be very similar)

I’m eager to hear what Mr. Martin has to say on this (he’s probably already got his slide rule out…) but I imagine we’d start by figuring that a given session knows, maybe, 300 tunes? I’m doing a bit of Fermiation, but here’s how I figure: a session might run four hours, not more. A set of tunes is about five minutes and has about 3 tunes in it. That’s 36 tunes (3X12) per hour, 144 tunes over the course of the night tops. (I’m leaving no time for craic here, of course, to get a maximal number)

If you assume that no tune is repeated in the course of a month, that’ll give you a bit under 600 tunes. But we know that there are tunes repeated from week to week, so 600 would be the high boundary of tunes that could be actually played at one session in a given month. That doesn’t tell you how many tunes might be known by a given player at that session if you were to drop in and start playing your favorite obscure slides out of Breathnach’s collections, but it does give us something concrete to play with.
I think it’s safe to say that a given session might play, at a maximum, 300 unique tunes over the course of a given month. Now, I may be making too much of a leap here but I imagine that the number of tunes played over the course of a given year will be very similar to that for a given month. At least, I can’t think of a whole lot of tunes that I play at a session less than once a month, so there’s not going to be a whole lot (in my thinking) to add to the monthly total.
So we can figure that about 300 more or less the stable repertoire of a given session.

Now of those, how many are unique to that session? Since we’re counting unique tunes played in a given month (or year), once we’ve counted the first session, we only need to count the tunes from the next session that we haven’t already played. So if we can figure how many tunes session number two is playing that session number one isn’t, we’re closer to a RG’s number. I don’t know how to go about estimating that number, but I think it’s obvious that it’ll be a diminishing return as we increase the number of sessions. The math is probably pretty straightforward to someone who isn’t me, but it seems to me that it’s hard to get this to come out at even 2000 tunes in circulation, that is, having been played in a session recently.

Of course, the number of places that I could have gone wrong in this calculation is so great that it’s statistically impossible I haven’t bollixed this up somewhere, so don’t be afraid to correct me where I’ve gone astray!

Re: How many tunes are there?

The compulsive side of me writes down all the tunes I’ve heard in Vancouver in hte past 2 years. It’s an underestimate because I am not perfectly diligent, and because frequently the player doesn’t know the name. I keep tunes played by visitors separate. And I am not counting the three people in town who have prodigious repertoires

Anyhow, the number is 500. FWIW, about 450 are on the lists compiled by Will Harmon. I interpret this to mean that Will has done a great job of providing a guestimate of tunes that are widely played. Even assuming there are a lot of local/regional tunes that aren’t on Will’s list.

I don’t think this changes anything in Jon’s calculation, so I agree with him that 2K is close.

Another way to do it would be to compare the repertoires of people who know many hundreds of tunes. I bet if you could get lists from ten people and made a metalist of unique tunes, you’d be close to the answer, and it would be in the 2-3K range.

Hugh

Re: How many tunes are there?

If n = number of tunes
p = number of people playing them
i = number of instruments being played by p
r/j = ratio of reels to (double) jigs included in n
s = number of sessions housing p
cd = number of recordings of n
f = frequency of tune repetition by p in s
g = inappropriate backing ( if this is >0 then p is reduced)
then, assuming h (number of decent hornpipes played) = 0
and allowing +/- 2% intolerance, and applying O’Carolan’s Algorithm as a limiting factor, we find that the mass of a tune increases in direct proportion to the speed at which it is played, leading to D (tune density) tending to 0 as we approach the event horizon (i.e., Banish Misfortune).

Or to put it another way, there aren’t as many tunes as there wouldn’t be if there weren’t any less.

I know I have.

Posted by .

Re: How many tunes are there?

Right. Perhaps the weakest part of the question is the definition of tune, but I am afraid that trying to make it any tighter could quickly degenerate into fisticuffs, although what I had in mind was pretty global, not referring to a given session, but a session anywhere, since this is an eminently global forum. Sounds like the original suggestion of thousands was not that far off the mark, if vague enough to be pretty safe.

Interesting to think that there might be some constraints on the number of "current" tunes that could effectively exist at any time, independently of the number of players or knowers of the tunes but determined by some other factor. I would argue for that number to be constantly increasing, given our obsession with "recording" our knowledge, although at an ever diminishing rate, given the progressive depletion of new possibilities.

Anyway, in the end, it is all about as meaningful as asking "what’s the name of that tune?", but sometimes it’s amusing to ponder puzzling little things. Another one of my favourites is the ancestor paradox. Anyone got any answers for that one?

Re: How many tunes are there?—Grossendeek’s Theorem

Found in GAG4, pp 8765:

The schemespec of diddly is right-left affronted. So…, the answer is 42.

The proof is 1482 pages long and uses all of the results from GAG1-GAG3.

Time to go learn that Martin Hayes jig I’ve been meaning to for the last two months…

Re: How many tunes are there?

Too bloody many

Posted .

Re: How many tunes are there?

But how many different tunes can all those instruments play? That’s what we want to know.
And just to mess up everyone’s calculations, don’t forget the trick notes, which increase the number of notes in a scale to 13 or 14, at least.

Re: How many tunes are there?

Looks like good prices on those saxophones - did they "fall
off the back of a truck" as they say in New Jersey?

Re: How many tunes are there?

more like fell off the back of a rickshaw πŸ™‚

Re: How many tunes are there?

There are two possible answers, depending whether you are a dedicated player of ITM or a novice listener:

1. An infinite number one will never be able to master;

or

2. One; they all sound alike.

Re: How many tunes are there?

Btw, for the avoidance of doubt, my previous post and the following ones by Hup and Bazouki Dave were our reactions to some troll’s post (now thankfully deleted) which was a commercial ad purporting to sell an enormous number of instruments, none of which had anything to do with ITM.

Re: How many tunes are there?

I run FolkTuneFinder.com, a site that indexes as many ABCs as I, and it, can find on the web. It’s not exclusively Irish and I know for a fact that it’s far from exhaustive (there are hundreds on waiting list but I don’t have the time just to fire up the search indexer just now).

But as you ask, you may be interested. It contains about 150,000 tunes. Abount 130,000 of those are unique (and the rest are duplicate transcriptions of each other). I’m sure if I had a scour, more could be found. My grab-a-number-out-of-the-air value for how many tunes that could eventually end up on it is 200,000, with 70% of those being unique transcriptions.

Joe