Makams Again

Sequel to Rationalizing 53-EDO for Turkish Makam Analysis and Quartertones (For Arabic Maqam Analysis).

I really want to figure out middle-eastern microtonal music. My last post was getting kind of long and busy, so I'll keep going here.

I once posted about the 24-EDO ("quartertone" analysis of Arabic maqamat. Here's a summary below. Each line has a maqam name, a number of steps of 24-EDO for each scale degree, a pitch class for each scale degree, and short comment about microtones. I just got these by transcribing staff notation off Wikipedia:

Hijaz (Nahawand ending) [0, 2, 8, 10, 14, 16, 20, 24] [D, Eb, F#, G, A, Bb, C, D] # Tonal. 

Nawa Athar [0, 4, 6, 12, 14, 16, 22, 24] [C, D, Eb, F#, G, Ab, B, C] # Tonal.

Shad 'Araban [0, 2, 8, 10, 14, 16, 22, 24] [G, Ab, B, C, D, Eb, F#, G] # Tonal.

Bayati [0, 3, 6, 10, 14, 16, 20, 24] [D, E-, F, G, A, Bb, C, D] # Has E-.

Jiharkah [0, 4, 8, 10, 14, 18, 21, 24] [F, G, A, Bb, C, D, E-, F] # Has E-.

Huzam [0, 3, 7, 9, 15, 17, 21, 24] [E-, F, G, Ab, B, C, D, E-]. # Has E-.

Rahat al-Arwah [0, 3, 7, 9, 15, 17, 21, 24] [B-, C, D, Eb, F#, G, A, B-] # Has B-.

Saba [0, 3, 6, 8, 14, 16, 20, 24] [D, E-, F, Gb, A, Bb, C, D] # Has E-.

Rast [0, 4, 7, 10, 14, 18, 21, 24] [C, D, E-, F, G, A, B-, C] # Has E- and B-.

Husayni 'Ushayran [0, 3, 6, 10, 13, 16, 20, 24] [A, B-, C, D, E-, F, G, A] # Has B- and E-.

Also, the maqamat 'Ajam, Nahawand, and Kurd are tonal, and correspond to western Major, Minor, and Phrygian modes. I don't know the traditional tonic pitches though, so they're not listed above.

Now, 24-EDO is kind of an ugly tuning system. There is no rank-2 interval that you can temper out to produce a 24-EDO tuning with the usual famous order of natural interval (P1 m2 M2 m3 M3 P4 P5 m6 M6 m7 M7 P8). That doesn't mean the scale doesn't exist: you can of course make music in it if you like, and other people can analyze it as being rank-3 or higher if they like. And honestly maqamat probably should be analyzed with 5-limit or higher prime frequency ratios. But I still think we can do better than 24-EDO as an analysis framework. Partly I think we can do better because Arabic, Turkish, and Persian maqamat/makams are supposedly developed from ancient Greek Pythagorean tuning, which had rank-2 microtones. And Turkish makams are analyzed with 53-EDO which is basically Pythagorean tuning, so rank-2 microtones not only should be enough theoretically, but they seem to be enough in practice. And I'm told that Arabic music isn't even played as 24-EDO, that's just a short-hand for notation which doesn't match the measured frequencies. ... So it's great that most of the pitches of the maqamat above are also natural pitches found in 12-EDO. Among all the maqamat above, we only have E- and B- as microtones. So let's try to analyze those in the same 53-EDO framework as the Turkish makams, rather than 24-EDO, which I dodn't care for. And then we'll compare the Arabic maqamat in this post to the Turkish makams in the previous post. And hopefully everything will come together beautifully. The information I have from one musical tradition will clear up the confusions I have about the other, and vice versa.

Let's start by writing the maqams above in terms of intervals, rather than tuned steps of an EDO or pitch classes.

Here's the Arabic version of Rast:
    ['C', 'D', 'E-', 'F', 'G', 'A', 'B-', 'C'] 
.
The Turkish version had scale degrees on [0, 9, 17, 22, 31, 40, 48, 53] of 53-EDO, which correspond to 

    [P1, M2, d4, P4, P5, M6, d8, P8]

as simple intervals. 

That scale is malformed numerically/alphabetically: we have a small fourth instead of a third interval, and a small octave instead of a seventh interval. But the scale is still nice in that it has diminished intervals where the Arabic notation suggests a microtone. Let's spell it in rank-2 pitches instead of using those weird +/- a quarter tone accidentals:

    [C, D, Fb, F, G, A, Cb, C]

In Pythagorean tuning, a {Cb} is a little bit below a {B}, so don't go thinking "Oh, Cb = B, but Rast has a B half flat, so this isn't flat enough." I'm here to tell you, {Cb} does not equal a {B}, and this has consequences for microtonal music.

If we go up to a rank-3 analysis, then Rast is 

    [P1, AcM2, M3, P4, P5, AcM6, M7, P8]

This is well-formed as a scale, but corresponds less well to the Arabic staff notation.

Let's try to find another microtonal maqam where we also have an analysis for a Turkish makam with approximately the same name.

Arabic Huzam and Turkish Huzzam are clearly a pair, but I was confused about Huzzam, so that's not a good test case for confirmation. I need to cement my understanding of Arabic maqamat before I use it to fix my understanding of Turkish makams.

Let's do Arabic Bayati versus Turkish Beyâti!

The ascending Turkish Beyâti is defined by these "simge":

[K S T T B T T]

which correspond to these numbers of "commas" or steps of 53-EDO:
     [8, 5, 9, 9, 4, 9, 9]

which we can accumulate in a running sum to get the tuned step sizes for each Beyâti scale degree:
 
[0, 8, 13, 22, 31, 35, 44, 53]

which correspond to these simple intervals:

[P1, d3, m3, P4, P5, m6, m7, P8]
.
We compare to the Arabic version above....
[D, E-, F, G, A, Bb, C, D] 

And it all works again! We have a diminished third instead of a second interval, as might have been predicted from the case with Rast, and all of the other pitch classes match what they should be: i.e. Bb really is a minor sixth above D. I'm so happy! The simple rank-2 analysis is working really well. I believe the Arabic Bayati here is specifically "Bayati with Nahawand ending" in contrast to "Bayati with Rast ending".

Now let's try using the Arabic Huzam and Saba to figure out the Turkish versions! Because I was super confused about those. The Arabic Saba is really close to the Arabic Bayati: the G on the fourth scale degree just becomes a Gb. And a Gb is a diminished fourth above D, so we have:

Arabic Saba: [P1, d3, m3, d4, P5, m6, m7, P8]

If we tune those rank-2 intervals in 53-EDO, we get these as the tuned steps of (Arabic) Saba:
[0, 8, 13, 17, 31, 35, 44, 53]
.
This almost matches a description of Saba from the 53-EDO / Turkish maqam post: Ali C. Gedik gave this description of Saba in a dissertation:

    [0, 8, 13, 18, 31, 35, 44, 49]

This differs on the fourth scale degree (where his written 18 is the number of 53-EDO steps for the tuned value for M3 not d4), and on the 8th scale degree (where his written 49 steps corresponds to a M7).

The 18 step 4th scale degree makes more sense in Turkish music theory than a step at 17 would: If we use 18, then

    31 - 18 = 13
    P5 - M3 = m3

we have a normal interval with a named simge ("A13") between the fourth and fifth scale degrees. If we use the 17, then the difference is 14 steps of 53-EDO, which Turkish music theory doesn't have a simge for.

Let's look at some different source for both the Arabic and the Turkish Saba. For Arabic, maqamworld.com gives Saba as either:

    Saba with 'Ajam ending: [D, Eb-, F, Gb, A, Bb, C, D]
    Saba with Nikriz ending: [D, Eb-, F, Gb, A, Bb, C, Db, E, F]

Excellent. I've seen some Turkish sources give an extended Saba that misses and goes over the octave as:

    [K S S A B T S A S]

The "A"s are a little ambiguous between 12 and 13 steps of 53-EDO in Turkish music theory, but it won't be a problem here: We'll use 13 steps for the first A, just like Gedik. Next, we're told from the Arabic Saba page that the extended form of Saba has a scale fragment (or "jins" (plural "ajnas") called Nikriz built on the sixth scale degree. The 5-note Nikriz pentachord spans a perfect fifth, and we need 12 steps (of Turkish 53-EDO) on the {A} for that to work out. Here's the Nikriz pentachord in simge:

    [T, S, A12, S]

and again in relative EDO steps from the previous scale degree 

    [0, 9, 5, 12, 5]

and here are the running totals:

    [0, 9, 14, 26, 31]
and how about the rank-2 intervals, why not?

    [P1, M2, A2, d5, P5]

With 12 steps on the second {A} simge, here are all the scale degrees in 53-EDO steps:

    [0, 8, 13, 18, 31, 35, 44, 49, 61, 66]

which correspond to these simple intervals:

    [P1, d3, m3, M3, P5, m6, m7, M7, d10, m10]

This is an extended Turkish Saba, perhaps, if I did everything right. And it doesn't match the Arabic version. The fourth scale degree is still mismatched, like before (18 for Arabic versus 17 for Turkish). But also the Arabic one ends "C, Db, E, F" and this guy above with the "d10" ends "C, C#, Fb, F" if you spell it out. Those Arabic pitches are the sorts of substitutions you might make from the Turkish scale if you work in 12-EDO or 24-EDO. But I wouldn't have minded if the two intervallic analyses had been identical.

I know that was a lot. But this is extended Saba as far as I can tell from Turkish notes on sigme:

    [P1, d3, m3, M3, P5, m6, m7, M7, d10, m10]

And this is extended Saba as far as I can tell from Arabic staff notation:

    [P1, d3, m3, d4, P5, m6, m7, d8, m9, m10]

Maybe the cultures actually have slightly different scales, or maybe the differences of staff notation and the EDOs they use for analysis are introducing noise into my intervallic analysis. I'm not sure. I think because Turkish music has a finer grained analysis, I tend to trust it more. And also, Arabic 24-EDO collapses lots of enharmonic interval distinctions, in addition to being coarser grained. So I'm leaning toward trusting the Turkish analysis, in so far as the two musical traditions share one object called Saba.

I was kind of hoping that any time an Nth scale degree had a diminished (N+1)th interval in the simge analysis, that the Arabic staff notation would notate a half-flat microtone, and vice versa. But the d10 at the end of the extended Turkish Saba didn't bear that out. So I don't have a general principle for converting between 24-EDO and 53-EDO notations.

Still this is good progress: in the last post, I was considering the possibility that the first two simge of the extended Turkish Saba were ornaments, and that [S A13 B T S A12 S] was the version that went from P1 to P8. Ridiculous! I'm learning a lot.

Ready to analyze Huzam/Huzzam? I know I am. 

The Arabic Huzam goes

    [E-, F, G, Ab, B, C, D, E-]

The Turkish Hüzzâm is notated

    [S T S A S A B]

in simge. The {A} simge is ambiguous between 12 and 13 microtones, but the assignment that makes the most sense is

    [S T S A12 S A13 B]

which becomes this

    [0, 5, 14, 19, 31, 36, 49, 53]

if you convert simge to EDO steps and take a running sum. Ali C. Gedik also gives those numbers exactly in his dissertation, so I've got multiple directions of confirmation that this is the Turkish Huzzam.

This one frustrated me because I'd heard that Turkish music only uses intervals tuned to these steps

    [0, 4, 5, 8, 9, 13, 14, 17, 18, 22, 23, 26, 27, 30, 31, 35, 36, 39, 40, 44, 45, 48, 49]

of 53-EDO, which doesn't include 19. Those scale degrees [0, 4, 5, 8, ...] also correspond to natural and once-modified intervals, whereas the simplest rank-2 interval that 53-EDO tunes to 19 steps is a twice augmented second, AA2. So 19 steps is a weird duck.

If we convert all of the Huzzam steps to simple rank-2 intervals, we get:
    
    [P1, A1, A2, AA2, P5, A5, M7, P8]

Pretty weird. In the quartertone post, I was tempted to say, "Maybe E- isn't the actual tonic. What happens if we permute the scale and pretend it starts on C?". I'm not sure if I'll do that here, but I'll remark that Huzam is not looking any less weird in Turkey than it did in Arabia.

If we convert the Huzzam intervals to pitches rooted on C (not permuting the scale to start on the sixth scale degree, just taking the interval [P1, A1, ...] and stacking them on top of a C), then we get

    [C, C#, D#, D##, G, G#, B, C]
    
which you might be tempted to spell enharmonically as

    [C, Db, Eb, Fb, G, Ab, B, C]

Even though that changes and ruins all the interval and only works in 12-TET. This is almost the form we see on Piano Encyclopedia's page on Huzzam in C

    [C, Db, Eb, Fb, G, Ab, Bb, C]

It only differs in making the {B} into a {Bb}. But multiple sources confirmed the intervals of Turkish Huzzam for me, so I think I trust my own {B} more than the Piano Encyclopedia's {Bb}. Although maybe they're giving me Arabic Huzam? Because they spell it "maqam" rather than "makam" on the page. I don't know, man.

The Arabic "Rahat al-Arwah" maqam listed at the start of the post has the same intervals as Huzam but it's rooted on B- instead of E-. They seems pretty serious over there in the middle east about liking this weird scale, and about rooting it on a microtone, even though they perhaps only have two microtones in total. But also, B- and E- aren't real pitches, they're just weird accidentals made up for analyzing things in 24-EDO, and I want to connect them to real rank-2 intervals and rank-2 pitches.

So here's what we're going to do: take the Turkish Huzzam and transpose it so that the fifth scale degree, which is P5 over the tonic in Turkey, matches the fifth scale degree of the Arabic Huzam rooted on E-. We'll do that transposition, and then we'll compare pitches.

The Arabic Huzam has a B natural for its fifth scale degree. Therefore we'll root the Turkish Huzam on E natural, since B is a perfect fifth above E natural. Our old wonky friend, the Turkish Huzzam,

    [P1, A1, A2, AA2, P5, A5, M7, P8]

becomes

    [E, E#, F##, F###, B, B#, D#, E]

which we shouldn't simplify by 12-EDO enharmonic respelling, but let's see how it goes anyway:

    [E, F, G, Ab, B, C, D#, E]

and now compare to the Arabic Huzam:

    [E-, F, G, Ab, B, C, D, E-]

The half flat microtones are obviously missing, which is sad. And also, this Huzam has a D instead of a D#, just like how Piano Encyclopedia used a m7 instead of a M7.

I'm getting sad. I want a professor of middle eastern music theory to explain it all away.

...

There aren't very many pitches in the maqamat I got from wikipedia. Saba is the only one with a Gb, and besides that there are only 12 pitch classes, not a full 24. This is the full set of 13:

    [A, Bb, B-, B, C, D, Eb, E-, F, F#, Gb, G, Ab]

and Gb is tuned the same as F# for them, so it's really just 12. They operate on this very cute little subspace of what their music theory allows.

...

Oh! Yeah, the respelling is what made me sad. The first interval of Huzam, at least in a rank-2 intervallic analysis, is an augmented unison. That's what it is in Turkey and Turks got the math right, or at least right enough for a very productive Pythagorean analysis. It makes sense that anyone who analyzes it with 24-EDO would spell it wrong. It's still a mystery to me why Arabic maqam seems to have a m7 and Turkish seems to have a M7, but maybe that's because they play different scales.

Anyway! When we compared the Arabic and Turkish descriptions of Rast, we concluded that {E-} is really an {Fb}. So let's spell Huzzam with an Fb for the root!

    [Fb, F, G, G#, Cb, C, Eb, Fb]

I think this significantly better than starting on E. It only introduces one new pitch class, G#, relative to the 13 pitch classes in Arabic maqamat (when they're spelled in the Pythagorean way with Cb and Fb instead of B- and E-). It's the way you'd spell you'd spell your maqam if you already had a 12-EDO lute with extra fretlets added in for Cb and Fb. And maybe Arabs play D instead of Eb for the 7th scale degree, I don't know.

Okay, what's left to analyze of the Arabic maqamat? Hijaz, Shad Araban, and Nawa Athar are all tonal. Here are some cute facts about them from the Quartertones post in the meantime:

    * If you spell Hijaz by thirds instead of by step, it's a D.11b9b13 chord, aka Phrygian with a major 3rd.
    * If you spell Shad Araban by thirds, it's a G.Maj11b9b13 as a chord. Also it's a permutation of Nawa Athar.
    * If you spell Nawa Athar by thirds, it's a C.minor-major9#11b13 chord.

I've heard that Jiharkah is uncommon in Arabic music. I don't know if it has a counterpart in Turkish music. And I still don't really have a general procedure for figuring out the actual rank-2 intervals that Arabic maqams notated in 24-EDO are made of.

Might the Arabic maqam Husayni Ushayran correspond to the Turkish makam Hüseynî? Let's find out!

The simge for Hüseynî are: 

    [K S T T K S T]

which in commas can be written

    [8, 5, 9, 9, 8, 5, 9]

Here are the running sums:
    
    [0, 8, 13, 22, 31, 39, 44, 53]

And here are simple corresponding intervals:

    [P1, d3, m3, P4, P5, d7, m7, P8]

Rooted on {A}, this becomes:

    [A, Cb, C, D, E, Gb, G, A]

So that's the spelling of the Turkish makam Hüseynî. Here's the Arabic maqam again, Husayni Ushayran, ([A, B-, C, D, E-, F, G, A]), spelled with rank-2 pitches:

    [A, Cb, C, D, Fb, F, G, A]
    
Close, but no cigar. It's cool that the Arabic one has a diminished sixth instead of a perfect fifth.

Oh, weird. Alsiadi.com gives a description of Maqam Husayni (the Arabic spelling), but relates the intervals by commas. But also, the commas don't make sense? There are 6s and 7s, which don't have simge in Turkish music theory. But also interestingly, the scale has the same accidentals as wikipedia's Husayni Ushayran, but roots the scale on D:

    [D, Fb, F, G, A, Cb, C, D]

and this is exactly the Turkish Huseyni transposed to D. So... the wikipedia Husayni is a permutation of the Alsiadi.com Husayni, which is incorrectly notated with commas, but if we ignore them then it's spelled the same as the Turkish makam Hüseynî.

I wish these things made more sense. I think I've also heard that Turkish music is played like P4 lower than notated? And {D} is P4 above {A}. So there.

The site "learnarabicmusic.com" also starts on D, with half flats on E (=Fb) and B (=Cb). So that's encouraging. I think wikipedia is just wrong.

...

I transcribed all but one of the maqamat on maqamworld.com. Some of them have notes about the ajnas (trichords, tetrachords, pentachords, hexachords) from which they're built. Maqamworld.com has more information; I just wasn't very thorough with it.

:: Arabic Maqamat:
'Ajam Family:
'Ajam (Upper Ajam Ending): [C, D, E, F, G, A, B, C] # 'Ajam pentachord + upper 'Ajam tetrachord. Major scale.
'Ajam (Nahawand Ending): [C, D, E, F, G, A, Bb, C] # 'Ajam pentachord + Nahawand tetrachord.
'Ajam 'Ushayran (descends): [Bb, A, G, F, Eb, D, C, Bb] # Nahawand trichord down to Kurd tetrachord down to 'Ajam tetrachord.
Shawq Afza: [C, D, E, F, G, Ab, B, C] # 'Ajam pentachord + Hijaz tetrachord
.
Bayati Family:
Bayati (Nahawand Ending): [D, E-, F, G, A, Bb, C, D]
Bayati (Rast Ending): [D, E-, F, G, A, B-, C, D]
Bayati Shuri: [D, E-, F, G, Ab, B, C, D]
Husayni: [D, E-, F, G, A, B- or Bb, C, D] # This has both descending and ascending parts, as written on MaqamWorld, and it has multiple sixth scale degrees. I don't get it. I've just written it ascending.
Muhayyar: Bayati (Rast Ending) but then you emphasize Jins Bayati on the octave?
.
Hijaz Family:
Hijaz (Nahawand Ending): [D, Eb, F#, G, A, Bb, C, D]
Hijaz (Rast Ending): [D, Eb, F#, G, A, B-, C, D]
Hijazkar (descends) : [E, Db, C, B, Ab, G, F, E, Db, C] # Also called Shadd 'Araban or Suzidil or Shahnaz
Zanjaran (descends): [C, Bb, A, G, F, E, Db, C]
.
Kurd Family:
Kurd: [D, Eb, F, G, A, Bb, C, D]
Hijazkar Kurd (descends): [E, Db, C, B or Bb, Ab, G, F, Eb, Db, C]
Nahawand Family:
Nahawand (Hijaz Ending): [C, D, Eb, F, G, Ab, B, C] # Nahawand pentachord + Hijaz tetrachord. Harmonic minor scale.
Nahawand (Kurd Ending): [C, D, Eb, F, G, Ab, Bb, C] # Nahawand pentachord + Kurd tetrachord. Natural minor scale.
Farahfaza: (Nahawand transposed to start on G.)
Nahawand Murassa': [C, D, Eb, F, Gb, A, Bb, C] # # Nahawand Murassa pentachord on the tonic, overlapped by Hijaz tetrachord on the 4th degree. And then add on the octave.
'Ushaq Masri: [D, E, F, G, A, B-, C, D] # Nahawand pentachord + Bayati tetrachord
.
Nikriz Family:
Nikriz (descends from 9 rather than octave): [D, C, Bb, A, G, F#, Eb, D, C]. Ascending, Nikriz pentachord + Nahawand pentachord.
Nawa Athar: [C, D, Eb, F, G, Ab, B, C] # Nikriz pentachord on tonic, overlapping with a Hijazkar Hexachord starting on third degree (centered on fifth degree)
Athar Kurd: [C, Db, Eb, F#, G, Ab, B, C]
.

Rast Family:
Rast (Upper Rast ending): [C, D, E-, F, G, A, B-, C]
Rast (Nahawand ending): [C, D, E-, F, G, A, Bb, C]
Kirdan: (descending Rast with upper rast ending)
Sazkar (descends): [C, B-, A, G, F, E-, D#, C]
Suznak: [C, D, E-, F, G, Ab, B, C] # Rast pentachord + Hijaz tetrachord.
Nairuz: [C, D, E-, F, G, A-, Bb, C] # Modern transposed maqam Yakah. Weird in that it has A half flat, which none of the other scales do.
Yakah: [G, A, B-, C, D, E-, F, G] # Older less common version of maqam Nairuz. Normal in that it has B- and E-, which many of the other scales do. Go team Yakah.
Dalanshin (descends): [E, Db, C, B-, A, G, F, E-, D, C]
Suzdalara (descends): [C, Bb, A, G, F, E-, D, C] # Just the descending form of rast with Nahawand ending.
Mahur: [C, D, E-, F, G, A, B, C]
.
Sikah family:
Sikah: [E-, F, G, A, B-, C, D, E-] # Sikah trichord + Upper Rast tetrachord + Rast trichord
Huzam: [E-, F, G, Ab, B, C, D, E-] # Sikah trichord + Hijaz tetrachord + rast trichord
Maqam Rahat al-Arwah: (Huzzam rooted on B-)
'Iraq: [B-, C, D, E-, F, G, A, B-] # Sika trichord + Bayati tetrachord + rast trichord
Awj ‘Iraq (descends): [D, C, B-, A#, G, F#, Eb, D, C, B-] 
Bastanikar: [B-, C, D, E-, F, Gb, A, Bb, C, Db, E, F] # "Maqam Bastanikar is effectively Jins Sikah followed by Maqam Saba. Its scale starts with the root Jins Sikah on the tonic, then Jins Saba on the 3rd degree, an overlapping Jins Hijaz on the 5th degree, and finally Jins Nikriz on the octave."
Musta'ar: [E-, F#, G, A, Bb, C, D, E-]

.
No family:
Jiharkah: [-E, F, G, A, B-, C, D, E-, F] # F is the tonic, I think, not E-. As much as Maqams have tonics, i.e. final notes. Jiharkah hexachord + Upper Rast tetrachord.
Lami: [D, Eb, F, G, Ab, Bb, C, D] # Lami pentachord, overlapped with kurd tetrachord on the fourth, then add on the octave.
Saba ('Ajam ending): [D, E-, F, Gb, A, Bb, C, D]
Saba (Nikriz ending): [D, E-, F, Gb, A, Bb, C, Db, E, F]
Saba Zamzam ('Ajam ending): [D, Eb, F, Gb, A, Bb, C, D]
Saba Zamzam (Nikriz ending): [D, Eb, F, Gb, A, Bb, C, Db, E, F]
.

The one maqam I didn't transcribe, Sikah Baladi, has weird accidentals that I don't know how to interpret. I think its the only one with weird intervals? There's a small chance I saw other weird flats and rounded them off in my head to "the half flat accidental".

Here's the comment on Sikah Baladi from maqam world: 

    "Maqam Sikah Baladi is arguably the most challenging Arabic maqam. Its scale (and sayr) is something of a hybrid between a transposition of Maqam Huzam to an ordinary non-Sikah note, and Maqam Hijazkar – the intervals are not quite the same as either, but it sounds a bit like both. None of its intervals match either just or equal-tempered intonation, making it impossible to reproduce on anything but the voice and traditional Arabic instruments.""

I don't believe for a fraction of a second that the intervals are neither just nor equally tempered. But it's cool that they recognize it's super duper not 24-EDO. Without knowing the staff notation, my attempt at a transcription would be something like:
Sikah Baladi (descends): [Cdown, Bup-, Adown-, G, Fup++, Edown-, D, C#, Cdown, Bup-, Adownb, G] "

I'll look into it and do it better. Or I'll find a different source on Sikah Baladi and talk about that characterization. 

Offtonic.com says: "An especially interesting variant, however, is Sikah Baladi. It results from exaggerated tuning of the Hijaz jins, so 1 b2 3 4 becomes 1 d2 d3 4. Instead of 5 - 12 - 5 commas, I used 6 - 10 - 6 commas in Sikah Baladi (53-TET) and 7 - 8 - 7 commas in Sikah Baladi X (53-TET), both in the Offtonic Scale Keyboard. Interestingly enough, the 24-TET values are sort of in between."

Eventually, I'll get all of the ajnas (trichords, tetrachords, pentachords, hexachords) written in, and that will speed along the process of figuring out fine-grained rank-2 intervals. But for now, I was really enjoying comparing the Turkish and Arabic relatives, and I might just find a few pairs to compare again.

I should also compare the MaqamWorld maqamat to the ones transcribed from wikipedia in case there are interesting differences. I've also found pitch classes for some Turkish makams on the Xenharmonic wiki. We'll synthesize and compare it all. Later.

From the names, I'm hopeful that we'll get at least 7 corresponding scales between the Arabic and the Turkish. And then maybe that will also give us Turkish intervals for some of the jins that make up the remaining Arabic scales.

:: Quick comparisons

: Segah with Sikah

The Turkish makam Segâh is defined by these simge:
[S, T, K, T, S, A13, B] : [5, 9, 8, 9, 5, 13, 4]

which we can accumulate to find these intervals:
[0, 5, 14, 22, 31, 36, 49, 53] :: [P1, A1, A2, P4, P5, A5, M7, P8]

Maybe Turkish Segâh is related to Arabic Sikah? Let's see what the Segah pitch-classes look like rooted on Fb, as the Arabic Sikah is rooted on E-:
    [Fb, F, G, Bbb, Cb, C, Eb, Fb]

and now compare to Sikah:
    Sikah: [E-, F, G, A, B-, C, D, E-]

Hm! The substitution of Turkish {Bbb} for Arabic {A} isn't too weird, with Arabs/24-EDO ignoring enharmonic differences. The replacement of {Cb} with {B-} is as expected. But we have a real difference with the Turks using {Eb} for the 7th scale degree and the Arabs using {D}.

: Nikriz with Nikriz
Turkish Nikriz is defined by these simge:
    [T, S, A12, S, T, K, S] : [9, 5, 12, 5, 9, 8, 5]

which we can accumulate to get these steps and simple intervals:
    [0, 9, 14, 26, 31, 40, 48, 53] :: [P1, M2, A2, d5, P5, M6, d8, P8]

The Arabic Nikriz has its tonic on C but descends from a D above the octave. Let's knock off the high D and write the whole arabic maqam as ascending, like the Turkish one:
    [C, D, Eb, F#, G, A, Bb, C]

If we root the Turkish Nikriz intervals on C, we get
    [C, D, D#, Gb, G, A, Cb, C]

The differences of Eb/D# and F#/Gb are to be expected enharmonically. The difference on the seventh scale degree seems genuine. Either I'm doing something wrong, or the two traditions use similar scales with differing seventh degrees. Segah/Sikah also differed on the seventh scale degree. Hm.

: Irak with 'Iraq

Turkish Irak is defined by these simge and relative steps:
    [S, T, K, S, T, T, K] : [5, 9, 8, 5, 9, 9, 8]

which we can accumulate to get these scale degree steps and simple intervals:
    [0, 5, 14, 22, 27, 36, 45, 53] :: [P1, A1, A2, P4, A4, A5, A6, P8]
.
Arabic 'Iraq is rooted on B half flat, 
    [B-, C, D, E-, F, G, A, B-]

which in my rank-2 analysis is called Cb. If we root the Turkish Irak on Cb, we get:
    [Cb, C, D, Fb, F, G, A, Cb]
    
Perfect agreement! I was starting to doubt myself.

: Hüseynî with Husayni

In my analysis of Turkish makams, I was under the impression that both Nevâ and Hüseynî were defined by these simge and relative steps in 53 EDO: 
    [K, S, T, T, K, S, T] : [8, 5, 9, 9, 8, 5, 9]

which can be accumulated to get these scale degree steps and simple intervals:
    [0, 8, 13, 22, 31, 39, 44, 53] :: [P1, d3, m3, P4, P5, d7, m7, P8]

Let's root the Turkish scale on D, like the Arabic maqam Husayni:
    [D, Fb, F, G, A, Cb, C, D]

and now we'll compare to the Arabic:
    [D, E-, F, G, A, B- or Bb, C, D]

Good agreement! On MaqamWorld, multiple pitches were listed for the 6th scale degrees of this maqam (B- or Bb). It seems Turkish music theory goes with the B- (i.e. Cb). The site alsiadi.com (with arabic maqam names ad pseudo-turkish commas) also lists B-.

: Sûzinâk with Suznak

The Turkish makam Basit Sûzinâk is defined by these simge and relative steps:
    [T, K, S, T, S, A12, S] : [9, 8, 5, 9, 5, 12, 5]

We can accumulate these give to give scale degree steps in 53-EDO, which are the tuned values for these simple rank-2 intervals: 
    [0, 9, 17, 22, 31, 36, 48, 53] :: [P1, M2, d4, P4, P5, A5, d8, P8]

The Arabic maqam Suznak is notated:
    [C, D, E-, F, G, Ab, B, C]

If we root the Turkish intervals on C, we get these pitch classes:
    [C, D, Fb, F, G, G#, Cb, C]

The differences on the third and sixth scale degrees are enharmonic. The seventh scale degree seems to be a real difference, as far as I can tell from notation. Arabs are playing B natural, and Turks are playing what Arabs would call B half-flat.

: Acem and Uşşak versus Ajam and 'Ushaq Masri?

As far as I could tell in my reading of Turkish makams, all of Acem and Uşşak and Beyâti were defined by the same simge. And we've already successfully analyzed Beyâti. But them Turkish "Acem" kind of sounds like Arabic "Ajam" and Turkish "Ussak" kind of sounds like arabic "'Ushaq". I'm not sure what do about those. I guess I'll have to figure out the Turkish intervals for the scale fragments (jins) that make up the Arabic maqams to find out the rank-2 intervals, because I don't have a full Turkish scale with which to compare. Maybe I could compare against theoretical Turkish scales, i.e. those made out of simge and tuned in 53-EDO with natural intervals, and see if any of them can be enharmonically respelled to match Arabic maqamat like 'Ushaq. 

I think in Arabic, 'Ajam means "mute", and is used to refer to people who don't speak their mother tongue, i.e. it's a pejoratives for foreigners. And in lots of other languages, including Turkish, it refers to Iranians. So maybe 'Ajam is a foreign scale for the Arabs and a native scale for the Turks or Persians with a different name. But .... 

: Çârgâh versus 'Ajam

If I just look at the pitches / intervals and not the makam names, the Turkish makam Çârgâh and the Arabic maqam 'Ajam are just the C major scale. Not even worth comparing. Interesting though that the Arabic major scale is called "damn foreigner". It makes me wonder if maqam Rast came first for them and how that happened and what came next.

If I had just been looking at the names of maqamat, I might have tried comparing Çârgâh to Jiharkah. No need for that.

: Nivahend versus Nahawand

In Turkish music, makam Nivahend and makam Buselik have the same intervals, just a different root position. The ascending form of the scale is defined in term of simge and relative steps in 53-EDO as:
    [T, B, T, T, B, T, T] : [9, 4, 9, 9, 4, 9, 9]

We can accumulate the steps to get absolute steps for each scale degree relative to the tonic. Here are also some simple intervals that 53-EDO tunes to each of those scale degrees:
    [0, 9, 13, 22, 31, 35, 44, 53] :: [P1, M2, m3, P4, P5, m6, m7, P8]

If we root those intervals on the pitch class C, then we get this scale of pitch classes:
    [C, D, Eb, F, G, Ab, Bb, C]

which exactly matches the Arabic maqam called Nahawand with Kurd ending. It's composed of a Nahawand pentachord and a Kurd tetrachord. It's also called the natural minor scale in western music theory. It's the type of minor scale that's a cyclic permutation of the major scale. Also called the Aeolian mode.

The harmonic minor scale
    [C, D, Eb, F, G, Ab, B, C]

has a related name in Arabic muisc theory: it's called Nahawand with Hijaz ending. The Nahawand maqam intervals rooted on {G} instead of {C} are called maqam Farahfaza, and I think again you can use either the harmonic minor ending (Hijaz tetrachord) or the natural minor ending (Kurd tetrachord).

...

: Mahur with Mahur

Mahur on Turkish music theory sites has the same commas as a major scale
    [T T B T T T B]

and is rooted on G and is written descending. But the staff notation  has a weird somewhat-sharp accidental (Küçük mücenneb?) on the F, but I don't even like Western staff notation and I can't even. I think it corresponds to the simge S, i.e. 5 commas sharp. But ...then why are the intervals just for a major scale? I don't know man. Mahur on MaqamWorld is like a major scale but with E half flat:
    [C, D, E-, F, G, A, B, C]

So maybe Mahur is a major scale with either a half flat third scale degree or a 5-comma sharp 7th scale degree.

:: Differences between Wikipedia's Maqamat and those of Maqam world:
: Nawa

Nawa Athar from wikipedia was:
[C, D, Eb, F#, G, Ab, B, C]

Nawa Athar from MaqamWorld:
[C, D, Eb, F, G, Ab, B, C]

They differ on the fourth scale degree. I trust the second source way more, but this is still something to investigate.

: Shadd

Shad 'Araban on Wikipedia was rooted on G and ascends:
    [G, Ab, B, C, D, Eb, F#, G]

Shadd 'Araban on MaqamWorld (also called Hijazkar or Suzidil or Shahnaz) was rooted on C and descends, but it descends from two scale degrees above the tonic:
    [E, Db, C, B, Ab, G, F, E, Db, C]

If we reverse that and start limit it to an octave,
    [C, Db, E, F, G, Ab, B, C]
 
Then .... yeah, those are the same intervals as on Wikipedia. 
    [P1, m2, M3, P4, P5, m6, M7, P8]

Nice.

: Jiharkah
Jiharkah on Wikipedia is 
    [F, G, A, Bb, C, D, E-, F]

while on MaqamWorld it is 
    [-E, F, G, A, B-, C, D, E-, F]

with whole notes on the Fs, suggesting to me that those have a tonic function. There's a difference of Bb versus B- on the fourth scale degree.

: Husayni 'Ushayran
I think the last difference is that MaqamWorld doesn't have Husayni 'Ushayran at all. The website alsiadi.com gives the same ascending version as Wikipedia, namely
    [A, B-, C, D, E-, F, G, A]

but then also lists a descending form, which I'll write ascending for ease of comparison:
       [A, B-, C, D, Eb, F#, G, A]
.

That's it for differences from Wikipedia.

I think next I want to ...do a 5-limit analysis of all the scales real quick? Yes, I do.

Ah, but alsiadi.com has maqamat that MaqamWorld doesn't. Maybe I should gobble those up real quick. It seems like the sort of site that could go down any minute. Also, he has weird 53-EDO steps that don't make sense in Turkish analysis. Maybe they're clues to the fine-grained differences in performance of Arab music relative to Turkish, clues we can't get from 24-EDO analysis or notation. And then, uh, there's this great youtube channel called Oud For Guitarists. I'll check if his descriptions of any of the makams differ. And then I'll have a really strong foundation in this stuff. Many sources, multiple cultures, independent partial confirmations or interesting contradictions. 

And then I'll do a five-limit analysis. And then I'll look at Ozan Yarman's weird, like, 23-limit frequency ratios, which I expect to be good (in their match to measured performances), even though the notated differences between frequency ratios are all sadly mathematically wrong. And I shouldn't be surprised, because anyone who spends too much time on the Xenharmonic wiki forgets how to do arithmetic.

...

Okay, from Mr. Alsiadi's site:

Rast ascending is the usual [C, D, E-, F, G, A, B-, C], i.e. the Upper Rast ending, but it's notated with commas [9, 7, 6, 9, 9, 7, 6]. # Rast tetrachord + 9/8 + Rast tetrachord.

Rast ascending for the Turks is analyzed with commas
    [T, K, S, T, T, K, S] : [9, 8, 5, 9, 9, 8, 5]

Relative to them, Alsiadi is flattening each the E and the B by an extra comma. I guess this is more...foreign, detuned, microtonal that I'm used to, so maybe that's good? It's weird at least. If you want Rast to sound weird, this guy's commas are for you. None of this "basically a major scale if you're tone deaf" crap. This one stabs.

Rast descending has the Nahawand ending: [C, D, E-, F, G, A, Bb, C]. Here are his commas, written ascending also: [9, 7, 6, 9, 9, 4, 9]. # Rast tetrachord + 9/8 + Nahawand tetrachord.

Maqam Basandida (ascending): [C, D, Eb, F#, G, A, Bb, C] : [9, 4, 14, 4, 9, 4, 9] # Nakriz or Nawa Athar pentachord + busalik tetrachord
Maqam Basandida (descending): [C, D, E-, F, G, A, B-, C] : [9, 7, 6, 9, 9, 7, 6] # Rast tetrachord + 9/8 + Rast tetrachord. These are the same pitch classes that Rast uses to ascend.
Maqam Dalansheen (ascending): [C, D, E-, F, G, A, B-, C, Db, E, F] : [9, 7, 6, 9, 9, 7, 6, 4, 14, 4]. # Compound maqam. It's like Rast with an ornamentation on top that derives from an overlap with Saba (starting on the A).
Maqam Dalansheen (descending): [C, D, E-, F, G, A, Bb, C, D, E-, F] : [9, 7, 6, 9, 9, 4, 9, 9, 7, 6].
Mahur (ascending): [C, D, E, F, G, A, B, C] : [9, 9, 4, 9, 9, 9, 4]
Mahur (descending): [C, D, E-, F, G, A, Bb, C] : [9, 7, 6, 9, 9, 4, 9].

The ascending form matches my understanding of Turkish Mahur as basically a major scale (but also they put a kücük mücenneb accidental on the 7th scale degree in the staff notation, which doesn't even show up in their comm-based intervallic analysis).

Neither of Alsiadi's Mahur maqamat match those matches the Mahur from MaqamWorld, which was [C, D, E-, F, G, A, B, C]), but Alsiadi's descending form is closer and MaqamWorld's mahur descended, so we're making progress. I'm glad I'm going through this site. This is pretty interesting. I might never nail this stuff down.

Maqam Nishaburk (ascending): [D, E, F+, G, A, B-, C, D] : [9, 7, 6, 9, 6, 7, 9].

Maqam Nishaburk (descending): [D, E, F+, G, A, Bb, C, D] : [9, 7, 6, 9, 4, 9, 9].

To be clear, I'm just reproducing verbatim the commas that Alsiadi lists. I don't think for a second that they actually produce the pitches of the scale.

I've read (in the table of contents for "Harmonic Secrets of Arabic Music Scales" by Cameron Powers) that Nishaburk is Nayruz/Nairuz rooted on G. In fact, let's look at all the transposed maqam names that Powers gives:
Nishabur: (Ajam on D with Nahawand)
Bayati Ushayran: (Bayatayn on A)
Busalik Ushayran: (Bayati on A)
Midmi: (C Hijaz Kar 2nd becomes Tonic on Db)
Zanjaran: (Hijaz on C with Ajam)
Nuhuft: (Huseyni on A)
Irak: (Huzam on B half-flat with Bayati)
Bastanikar: (Huzam on B half-flat with Saba)
Rahat el Arwah: (Huzam on B half-flat)
Zirgulah: (Jaharka on D)
Shawki Tarab: (Kurd on A with Saba)
Tarz Nawin: (Kurd on C with Hijaz)
Hijaz Kar Kurd: (Kurd on C)
Farahfaza: (Nahawand 1 on G)
Sultani Yaka: (Nahawand 2 on G)
Ushaq Masri: (Nahawand on D with Bayati)
Nahawand Kurdi: (Nahawand on D)
Shiar: (Nahawand on E with Bayati)
Busalik: (Nahawand on E)
Dilkashidah: (Nahawand on G with Bayati)
Hisar: (Nawa Athar on D)
Nishaburk: (Nayruz on D)
Yekah: (Nayruz on G)
Yak-Gah (Rast on G): (Rast Nawa)
Qatar: (Saba Zamzamah on E)
Suznal: (Shawq Afza on C)
Panjigah: (Shawqi Awir on C)
Suzidil: (Shehnaz on A)
Jaharka Turki: (Shehnaz on F)
Sikah Balady: (Shehnaz on G with the “old intervals”)
Shad Araban: (Shehnaz on G)
Hijazi Ushayran: (Shuri on A)
Farahnak: (Sikah on B half-flat)
.
Nice list.

Okay, back to Alsiadi:

Maqam Suzdilara ascending (Jaharka ending): [C, D, E, F, G, A, Bb, C] : [9, 9, 4, 9, 9, 4, 9]
Maqam Suzdilara ascending (Bayati ending): [C, D, E, F, G, A B-, C] : [9, 9, 4, 9, 9, 6, 7]
Maqam Suzdilara descending: [C, D, E-, F, G, A, Bb, C] : [9, 7, 6, 9, 9, 4, 9] # The scale has Bb written, but then there are accidentals after the scale in the style of a key signature which has a half flat on the B line. So maybe you can do Bb or B- descending? But the notated jins above the scale says Jaharkah, which wouldn't gvie a microtone. Whatever. These scales are all the same. This is just Rast descending with the Nahawand ending.
Maqam Suznak ascending: [C, D, E-, F, G, A-, B, C] : [9, 7, 6, 9, 5, 13, 4]
Maqam Suznak descending: [C, D, E-, F, G, A, Bb, C] : [9, 7, 6, 9, 9, 4, 9] # Still just Rast descending with Nahawand ending.
Maqam Yakah: [G, A, B-, C, D, E-, F, G] : [9, 7, 6, 9, 6, 7, 9]
Maqam Nahawand (ascending): [C, D, Eb, F, G, Ab, Bb, C] : [9, 4, 9, 9, 4, 9, 9]
Maqam Nahawand (descending): [C, D, Eb, F, G, Ab, B, C] : [9, 4, 9, 9, 4, 14, 4] # this is the Hijaz ending for Nahawand per MaqamWorld
Maqam Al-Sinbulah (ascending) / Nahawand Murassah (ascending): [C, D, Eb, F, Gb, A, Bb, C] : [9, 4, 9, 4, 14, 4, 9]
Maqam Al-Sinbulah (descending) / Nahawand Murassah (descending): [C, D, Eb, F, G, Ab, B, C] : [9, 4, 9, 9, 4, 14, 4]
Maqam Farah Fazah (ascending): [G, A, Bb, C, D, Eb, F, G] : [9, 4, 9, 9, 4, 9, 9]
Maqam Farah Fazah (descending): [G, A, Bb, C, D, Eb, F#, G] : [9, 4, 9, 9, 4, 14, 4]
Maqam Hisar: [D, E-, F, G#, A, Bb, C#, D] : [6, 7, 14, 4, 4, 14, 4]
Maqam Nahawand Kabir (ascending): [C, D, Eb, F, G, Ab, B, C] : [9, 4, 9, 9, 4, 14, 4]
Maqam Nahawand Kabir (descending): [C, D, Eb, F, G, A, Bb, C] : [9, 4, 9, 9, 9, 4, 9]
Maqam Nakriz (ascending): [C, D, Eb, F#, G, A, Bb, C] : [9, 4, 14, 4, 9, 4, 9]
Maqam Nakriz (descending): [C, D, Eb, F#, G, A, B-, C] : [9, 4, 14, 4, 9, 7, 6]
Maqam Nawa Athar (ascending): [C, D, Eb, F#, G, Ab, B, C] : [9, 4, 14, 4, 4, 14, 4]
Maqam Nawa Athar (descending): [C, D, Eb, F#, G, Ab, Bb, C] : [9, 4, 14, 4, 4, 9, 9]
Maqam Sultani Yakah: [G, A, Bb, C, D, Eb, F#, G] : [9, 4, 9, 9, 4, 14, 4]
Maqam Hijaz Kar Kurd: [C, Db, Eb, F, G, Ab, Bb, C] : [4, 9, 9, 9, 4, 9, 9]
Maqam Ajam Kurdi: (page under construction)
Maqam Shawq Tarab (ascending): [A, Bb, C, D, E-, F, Gb, A] : [4, 9, 9, 6, 7, 4, 14]
Maqam Shawq Tarab (descending): [A, Bb, C, D, Eb, F, G, A] : [4, 9, 9, 4, 9, 9, 9]
Maqam Kurdi: [D, Eb, F, G, A, Bb, C, D] : [4, 9, 9, 9, 4, 9, 9]
Maqam Tarz Nawayn: [C, Db, Eb, F, Gb, A, Bb, C] : [4, 9, 9, 4, 14, 4, 9]
Maqam Hijaz (ascending): [D, E-, F#, G, A, B-, C, D] : [5, 13, 4, 9, 7, 6, 9] # Hijaz tetrachord + Rast pentachord
Maqam Hijaz (descending): [D, E-, F#, G, A, Bb, C#, D] : [5, 13, 4, 9, 4, 14, 4] # Hijaz tetrachord + 9/8 + Hijaz tetrachord
Maqam Hijaz Kar (ascending): [C, Db, E, F, G, Ab, B, C] : [4, 14, 4, 9, 4, 14, 4] # Hijaz tetrachord + 9/8 + Hijaz tetrachord
Maqam Hijaz Kar (descending): [C, Db, E, F, G, Ab, Bb, C] : [4, 14, 4, 9, 4, 9, 9] # Hijaz tetrachord + 9/8 + Kurd tetrachord. Alsiadi h as "Busalik tetrachord + 9/8 + Kurd tetrachord" notated in text, but the commas and pitch classes do not support that.
Maqam Shad Araban (ascending): [G, A-, B, C, D, Eb, F#, G] : [5, 13, 4, 9, 4, 14, 4] # Hijaz tetrachord + Nakriz pentachord. This one is really weird. The  pitch classes suggest a Nakriz pentachord [9, 4, 14, 4], but the commas written in originally were [9, 7, 6, 9] for the pentachord, which is Rast pentachord. Since the pitch classes and the pentachord label match each other, and the commas don't have any corroboration, I changed the commas for consistency. My version with the changed commas has the same intervals as Maqam Hijaz, but rooted on G instead of D. I've read on Oud For Guitarists that "maqams Hijaz kar, Shad Araban, Suzidil, and Shahnaz"  have the same intervals, so maybe that's encouraging.
Maqam Shad Araban (descending): [G, A-, B, C, D, Eb, F, G] : [5, 13, 4, 9, 4, 9, 9] # Hijaz tetrachord + 9/8 + Nahawand tetrachord
Maqam Shahnaz: [D, Eb, F#, G, A, Bb, C#, D] : [4, 14, 4, 9, 4, 14, 4] # Hijaz tetrachord + 9/8 + Hijaz tetrachord
Maqam Suzdal: [A, Bb, C#, D, E, F, G#, A] : [4, 14, 4, 9, 4, 14, 4] # Hijaz tetrachord + Nakriz pentachord
Maqam Zinkulah: [C, Db, E, F, G, A, Bb, C] : [4, 14, 4, 9, 9, 4, 9] # Hijaz tetrachord + Ajam pentachord

...

I got bored with that an transcribed another source. These are 
"The modal system of Arabian and Persian music" by Owen Wright, which mostly translates and analyzes the medieval music theory treatise Kitab al-Adwār by Safi al-Din al-Urmawi a.k.a Safiaddin Ormavi.

: maqmat
rahawi: [G, A-, B(-), C, D-, Eb, F, G]
'ushshaq : [G, A, B, C, D, E, F, G]
busalik: [G, Ab, Bb, C, Db, Eb, F, G]
nawa: [G, A, Bb, C, D, Eb, F, G]
rast: [G, A, B-, C, D, E-, F, G]
hijazi1: [G, A, B- or B(-), C, D-, E(-), F, G]
hijazi2: [G, A-, Bb, C, D-, E(-), F, G]
'iraq: [G, A-, B-, C, D-, E-, F, F#, G]
husayni: [G, A-, Bb, C, D-, Eb, F, G]
kardaniya: [G, A, B-, C, C#, D, E, F+, G]
buzurg: [G, A-, B(-), C, C#, D, (E, F+, G)]
zankula: [G, A, B-, C, D-, E(-), F, ((F#), G)]
zirafkand: [G, A-, Bb, C, D-, Eb, E-, F#, G]
kawasht: [G, A-, B-, C, D-, Eb, E-, (F#, G)]
muhayyir husayni: [G, A-, Bb, C, D, E-, F, G]
nihuft: [G, A-, B(-), C, D, E-, F, G]
ishfahan: [G, A, B-, C, D, E-, F, F#, G]

I don't know what the parentheses mean. I missed that when skimming the text on multiple passes. I think some of those might actually be medieval forms of the scales which have since been modified by history.

: ajnas

The three diatonic tetrachords:
/1 2 3 4/ # 'ushshaq
/1 2 3b 4/ # nawa
/1 2b 3b 4/ # busalik
And the three four-note zalzalian tetrachords:
/1 2 3- 4/ # rast
/1 2- 3- 4/ # iraq
/1 2- 3b 4/ # nawruz

Main form of hijazi tetrachord:
/1 2- 3 4/
Alternative form of hijazi tetrachord:
/1 2b 3 4/
Maybe also a possible alternative form of hijazi tetrachord?:
/1 2b 3- 4/
.

Normal pentachords:
/1 2- 3b 4 5/ # nawruz tetrachord + whole tone = husayni pentachord
/1 2 3b 4 5/ # rast tetrachord + whole tone = rast pentachord
/1 2- 3 4 5/ # hijazi tetrachord + whole tone = 'uzzal pentachord
/1 2b 3b 4 5/ # busalik tetrachord + whole tone = busalik pentachord
/1 2- 3- 4 5 / # iraq tetrachord + whole tone = iraq pentachord
/1 2 3 4 5 / # 'ushshaq tetrachord + whole tone = 'ushshaq pentachord

Weird ajnas, some of which are pentachords:
/1 2- 3b 3 4/ # isfahan
/1 2- 3 4 4# 5/ # buzurg
/1 2- 3b 3-/ # kuchek
/1 2- 3b 3- 4# 5/ # hisar
/1 2 3b 4 5/ # nawa tetrachord + whole tone = nawa pentachord
/1 2- 3b 3/ # rahawi

Hijazi might have been varied ascending and descending? Like this is a possibility
hijazi ascending:
/1 2- 3 4/

hijazi descending:
/4 3- 2b 1/

Okay, back to alsiadi:
...
Maqam Bayati ascending: [D, E-, F, G, A, B-, C, D] : [6, 7, 9, 9, 6, 7, 9] # Bayati tetrachord + 9/8 + Bayati tetrachord
Maqam Bayati descending: [D, E-, F, G, A, Bb, C, D] : [6, 7, 9, 9, 4, 9, 9] # Bayati tetrachord + 9/8 + Kurd tetrachord
Maqam Qarjighar (Bayati Shuri) (ascending): [D, E-, F, G, Ab, B, C, D] : [6, 7, 9, 4, 14, 4, 9] # Bayati tetrachord + hijaz pentachord
Maqam Qarjighar (Bayati Shuri) (descending): [D, E-, F, G, A, Bb, C, D] : [6, 7, 9, 9, 4, 9, 9] # Bayati tetrachord + busalik pentachord
Maqam Husayni (ascending): [D, E-, F, G, A, B-, C, D] : [6, 7, 9, 9, 6, 7, 9] # Bayati pentachord + Bayati tetrachord
Maqam Husayni (descending): [D, E-, F, G, A, Bb, C, D] : [6, 7, 9, 9, 4, 9, 9] # Bayati pentachord + Kurd tetrachord
Maqam Bayati Ushayran (ascending): [A, B-, C, D, E-, F, G, A] : [6, 7, 9, 6, 7, 9, 9] # Bayati tetrachord + Bayati pentachord
Maqam Bayati Ushayran (descending): [A, B-, C, D, E, F, G, A] : [6, 7, 9, 9, 4, 9, 9] # Bayati tetrachord + Busalik pentachord
Maqam Husayni Ushayran (ascending): [A, B-, C, D, E-, F, G, A] : [6, 7, 9, 6, 7, 9, 9] # Bayati tetrachord + Bayati pentachord
Maqam Husayni Ushayran (descending): [A, B-, C, D, Eb, F#, G, A] : [6, 7, 9, 4, 14, 4, 9] # Bayati tetrachord overlapping with Nakriz hexachord
Maqam Saba (ascending): [D, E-, F, Gb, A, Bb, C, Db] : [6, 7, 4, 14, 4, 9, 4] # Alsiadi describes this as having a Bayati trichord on "Re", a Hijaz tetrachord starting on "Fa", and a Hijaz trichord starting on the high "Do". As notated, the "Re" is the low "D" note though, as though C is the tonic of the scale. Also there is no third note written above the Db to complete the Hijaz trichord: there's just blank space. I don't know what a Hijaz trichord is in order to infer the pitch above "Db" or the simge-comma-integer after "4". Perhaps 14?
Maqam Saba (descending): [D, E-, F, Gb, A, Bb, C, D] : [6, 7, 4, 14, 4, 9, 9] # Bayati trichord + Hijaz tetrachord + Ajam trichord.
Maqam Saba Zamzamah: [D, Eb, F, Gb, A, Bb, C, D] : [4, 9, 4, 14, 4, 9, 9] # Kurd trichord + Hijaz tetrachord + Ajam trichord
Maqam Sikah (ascending): [E-, F, G, A, B-, C, D, E-] : [6, 9, 9, 7, 6, 9, 7] # Sikah trichord + Rast tetrachord + Rast trichord
Maqam Sikah (descending): [E-, F, G, A, Bb, C, D, E-] : [6, 9, 9, 4, 9, 9, 7] # Sikah trichord + Nahawand tetrachord + Rast trichord. Alsiadi's comments make it seem like he thinks the tonic is C rather than E-. That's fine. He can think what he likes.
Maqam Huzam: [E-, F, G, Ab, B, C, D, E-] : [6, 9, 4, 14, 4, 9, 7] # Sikah trichord + Hijaz tetrachord + Rast trichord.
Maqam Mayah: [E-, F, G, A, Bb, C, D, E-] : [6, 9, 9, 4, 9, 9, 7] # Sikah trichord + Nahawand tetrachord + Rast trichord
Maqam Musta'ar: [E-, F#, G, A, Bb, C, D, E-] : [11, 4, 9, 4, 9, 9, 7] # Musta'ar trichord + Nahawand tetrachord + Rast trichord
Maqam Jaharkah: [F, G, A, Bb, C, D, E-, F] : [9, 9, 4, 9, 9, 7, 6] # Jaharkah tetrachord + whole tone + Rast tetrachord
Maqam Ajam Ushayran: [Bb, C, D, Eb, F, G, A, Bb] : [9, 9, 4, 9, 9, 9, 4] # Ajam tetrachord + whole tone + Ajam tetrachord
Maqam Shawq Afza (ascending): [Bb, C, D, E-, F, Gb, A, Bb] : [9, 9, 6, 7, 4, 14, 4] # Ajam trichord (9, 9) + Saba pentachord (6, 7, 4, 14) overlapping with Hijaz tetrachord (4, 14, 4).
Maqam Shawq Afza (descending):  [Bb, C, D, Eb, F, G, A, Bb] : [9, 9, 4, 9, 9, 9, 4] # Ajam tetrachord (9, 9, 4) + whole tone + Jaharkah tetrachord (9, 9, 4).
Maqam Iraq (ascending): [B-, C, D, E-, F, G, A, B-] : [6, 9, 6, 7, 9, 9, 7] # Iraq trichord (6, 9) + Bayati tetrachord (6, 7, 9) + Rast trichord (9, 7)
Maqam Iraq (descending): [B-, C, D, E-, F, G, A, Bb] : [6, 9, 6, 7, 9, 9, 4] # Iraq trichord (6, 9) + Bayati tetrachord (6, 7, 9) + Busalik trichord (9, 4). This one has B- at the bottom and Bb at the top, and it doesn't form an octave. Really weird.
Maqam Awj Ara: [B-, C, D#, E-, F#, G, A#, B-] : [6, 14, 2, 11, 4, 14, 2] # Awj tetrachord (6, 14, 2) + Mustaar pentachord (11, 4, 14, 2)
Maqam Bastah Nikar: [B-, C, D, E-, F, Gb, A, Bb] : [7, 9, 7, 7, 4, 14, 4] # Iraq trichord (7, 9) + Saba tetrachord (7, 7, 4) + 14 commas + Kurd dichord (4).
Maqam Farahnak (ascending): [B-, C, D, E, F+, G, A, B-] : [6, 9, 9, 7, 6, 9, 7] # Sikah trichord (6, 9) + Rast tetrachord (9, 7, 6) + Rast trichord (9, 7)
Maqam Farahnak (descending): [B-, C, D, E, F+, G, A, Bb] : [6, 9, 6, 7, 6, 9, 4] # Sikah trichord (6, 9) + Rast tetrachord (9, 7, 6) + Busalik trichord (9, 4). This one has both B- and Bb and it doesn't form an octave. Weird.
Maqam Rahit Al Arwah (ascending): [B-, C, D, Eb, F#, G, A, B-] : [6, 9, 4, 14, 4, 9, 7] # Iraq trichord (6, 9) + Hijaz tetrachord (4, 14, 4) + Rast Trichord (9, 7)
Maqam Rahit Al Arwah (descending): [B-, C, D, Eb, F#, G, A, Bb] : [6, 9, 4, 14, 6, 9, 4] # Iraq trichord (6, 9) + Hijaz tetrachord (4, 14, 4) + Busalik Trichord (9, 4). Alsiadi has the Hijaz tetrachord annotated as [4, 14, 6] for the descending version, but that's not consistent with his pitch classes or how he quantifies Hijaz anywhere else, so I fixed it.

...

Done! Whew. That took several weeks because I didn't enjoy it. 74 maqamat and one "404, file not found". I bet I made a few error of transcription. Oh well! Time to analyze.

There are actually only unique 48 scales in terms of commas! I bet some are transpositions of each other and some descend the same way but ascend differently. If I remove the ones that say "(descending)", then there are still only 33 unique scales. 13 ascending scales have repeated commas, I guess.

Almost all of them sum to an octave with 53 commas. The ones that don't are:
75 commas - Maqam Dalansheen (ascending)
75 commas - Maqam Dalansheen (descending)
48 commas - Maqam Saba (ascending)
57 commas - Maqam Saba Zamzamah
50 commas - Maqam Iraq (descending)
52 commas - Maqam Bastah Nikar
47 commas - Maqam Farahnak (descending)
52 commas - Maqam Rahit Al Arwah (descending)

Bastah Nikar is only off by one comma. I double checked the text. It wasn't my mistake. I'd already commented about "Farahnak (descending)" and "Rahit Al Arwah (descending)" and "Iraq (descending)" not being octaves. Saba and Dalansheen are weird compound scales, no surprise there. I only found one typo this way, and I've already fixed it above. You'll never know what I got wrong.

Maqam Musta'ar and Maqam Awj Ara are the only ones with intervals of 11 commas.

If I look at cumulative intervals relative to the tonic, Maqam Bastah Nikar is the only one that hits 23 commas (though 22 and 24 are multiply attested across other maqamat, so I expect an off-by-one error somewhere in alsiadi's commas). Bastah Nikar is also the only one that reaches 30 commas (though 31 is dirty common). And Maqam Bastah Nikar is the only maqam to hit 7 commas. If we compare to other sources, "Bastanikar" on MaqamWorld is clearly a non-octave compound scale, so the "52" could be right. I don't know about the other relative and cumulative commas though. Still seems kind of suspicious.

"Maqam Suznak (ascending)" is the only maqam that reaches 36 commas, while 35 commas is really common. I think it might be okay though.

"Maqam Farahnak (descending)" is the only one that reaches 43 commas (and 44 is very common and 42 is unheard of). I tried a description of Farahnak from another source and got 1) a huge pdf of turkish and Arabic makams 2) nothing else. In a pdf with the file name "Music Theory Of Makams" and an internally listed title of "Turkish & Arab Makams: Music Theory For Oud", Farahnak is listed as ascending: 
    [B-, C, D, E, F#, G, A, B-, C, D, E-, F, G, A]. 

I believe the pdf was written by David Parfitt and originally hosted on oud.eclipse.co.uk. Anyway, Farahnak in this source is made of Sikah tetrachord overlapping by one note with Ajam pentachord overlapping by one note with Rast pentachord + Bayati pentachord. *This* is the sort of thing I expect from real maqam music. There's no such thing as trichords and dichords. This is a Farahnak I can believe in. The same source descends as:
    [B-, C, D, E, F#, G, A, Bb, C, D, E-, F, G, A]

i.e. one B- becomes a Bb in the middle. This lets you have a nihawand pentachord over (G, A, Bb, C, D). The same source lists two alternative forms of descending including the Rast pentachord.
    [..., G, A, B-, C, D, ...]
or 
    [..., D, E, F+, G, A, ...]
.
This an amazing book and I'm really sad I wasted so much time learning about makams from bad sources when this exists.

"Maqam Iraq (descending)" is the only maqam of Alsiadi's to hit 50 commas (and it lands on 50 commas). 

First I looked at the scales in a rank-2 analysis. And specifically at the ones that were enumerated / spelled correctly (in the sense of having a 1st interval, a 2nd interval, a 3rd interval, ...).

I found that "Ajam Ushayran" and "Mahur (ascending)" were both just the major scale: [P1, M2, M3, P4, P5, M6, M7, P8]

I found that "Suzdilara ascending (Jaharka ending)" is a cyclic permuation of the major scale, namely the Mixolydian mode: [P1, M2, M3, P4, P5, M6, m7, P8]

"Basandida (ascending)" and "Nakriz (ascending)" have the same intervals: [P1, M2, m3, A4, P5, M6, m7, P8] ## Nevermind, I had a typo. I don't know what Basandida (ascending) is.

Maqam Nawa Athar (ascending) and (descending) are both tonal
    [P1, M2, m3, A4, P5, m6, M7, P8] (ascending)
    [P1, M2, m3, A4, P5, m6, m7, P8] (descending)

"Nahawand Kabir (descending)" is the Dorian mode: [P1, M2, m3, P4, P5, M6, m7, P8]

Five different maqamat are just the harmonic minor scale (i.e. the Aeolian mode with a major seventh instead of a minor seventh). They are 
    "Al-Sinbulah (descending) / Nahawand Murassah (descending)", 
    "Farah Fazah (descending)", 
    "Nahawand (descending)", 
    "Nahawand Kabir (ascending)", 
    "Sultani Yakah".

And the intervals are: [P1, M2, m3, P4, P5, m6, M7, P8]

"Farah Fazah (ascending)" and "Nahawand (ascending)" are just the Aeolian mode: [P1, M2, m3, P4, P5, m6, m7, P8] .


"Hijaz Kar Kurd" and "Kurdi" are both Phrygian mode: [P1, m2, m3, P4, P5, m6, m7, P8] 

"Shawq Tarab (descending)" is the Locrian mode: [P1, m2, m3, P4, d5, m6, m7, P8]  

The Lydian mode is not represented. 

These three have the same intervals: "Maqam Hijaz Kar (ascending)", "Shahnaz" and "Suzdal". I don't know a Western name for the scale, but it is: [P1, m2, M3, P4, P5, m6, M7, P8]  

And these are all distinct and I don't know Western names for the scales:
    [P1, M2, m3, P4, d5, M6, m7, P8] Maqam Al-Sinbulah (ascending) / Nahawand Murassah (ascending) # Disregard this. I had a typo. I don't know what Al-Sibulah (ascending) is.
    [P1, m2, M3, P4, P5, M6, m7, P8] Maqam Zinkulah
    [P1, m2, M3, P4, P5, m6, m7, P8] Maqam Hijaz Kar (descending)
    [P1, m2, m3, P4, d5, M6, m7, P8] Maqam Tarz Nawayn
.

I don't think there's a way to specify a scale in terms of steps of 53-EDO and not have it possible to be enumerated / spelled correctly with rank 3 intervals. There is some rank3 1st interval on each step and some rank3 2nd interval on each step and so on. They might have very complex names, but there will be some analysis available.

I started out with a very simple analysis wherein every step of 53-EDO had at most two very simple candidate intervallic interpretations. This did well for most of the maqamat, but struggled to analyze 
* 17 steps as a kind of 4th interval (in Saba). Alsiadi's letter names indicate Gb (a diminished fourth above the tonic of D) but the rank-3 version of d4 is tuned to 19 steps in 53-EDO, not 17 steps. 17 steps is a major third and I didn't feel the need to consider any other interpretations. 
* 27 steps as a kind of 4th interval (in basandida ascending, and in Hisar, and in Nakriz ascending and descending, and in Nawa Athar ascending and descending). Maybe I could revisit this one. I considered 27-steps to be a Grd5. Here are all the weird 4th intervals that were nearby in my rank-3 analysis:
    
18: ["Grd4"],
19: ["d4"],
20: ["Acd4"],
21: ["Gr4"],
22: ["P4"],
23: ["Ac4"],
24: ["GrA4"],
25: ["A4"],
26: ["AcA4"],
So none of those interval names are options and I'd have to find something even weirder that's tuned to 27-steps. GrAA4?

* 26 steps as a kind of 5th interval (in Al-Sinbulah/Murassah ascending, and Shaq Tarab descending and Tarz Nawayn and Qarjighar (Bayati Shuri) (ascending) and Husayni Ushayran (descending).
* 39 steps as a kind of 7th interval in Shawq Tarab (ascending).
...

Interesting news! At least among maqamat starting on C, Alsiadi seems to assign pitch classes to 53-EDO steps pretty consistently.

C: P1 # 1/1
Db: Grm2 # 256/243
D: AcM2 # 9/8
Eb: Grm3 # 32/27
E-: GrM3 # 100/81
E: AcM3 # 81/64
F: P4 # 4/3
Gb: 26 steps of 53-EDO
F#: 27 steps of 53-EDO
G: P5 # 3/2
Ab: Grm6 # 128/81
A: AcM6 # 27/16
Bb: Grm7 # 16/9
B-: GrM7 # 50/27
B: AcM7 # 243/128
C: P8 # 2/1

I've written in 5-limit frequency ratios that are justly associated with the rank-3 intervals I'm analyzing his scale steps as being. Using those intervals doesn't produce the letter names that Alsiadi gives, so the table above is a portrait of his letter assignment, not a fact of 5-limit just intonation rooted on C. The frequency ratios are all Pythagorean except for the half flat tones:
E-: GrM3 # 100/81
B-: GrM7 # 50/27

and the two scale degrees that I didn't have an analysis for:
Gb: 26 steps of 53-EDO
F#: 27 steps of 53-EDO

Alsiadi's Mahur (ascending) was just a C major scale without microtones. I wonder why that didn't show up in the rank-2 analysis if all its frequency ratios are Pythagorean.

Let's look a little more at the half-flat tones. They are each two commas flatter than their natural pitch classes, E and B respectively. In a rank-2 analysis, E is 18 steps of 53-EDO over C and the interval is called M3. That's what Alsiadi appears to be doing. Alsiadi assigns the pitch class E- to 16 steps of 53-EDO, which in a rank-2 analysis is a ddd5. That has a pitch class of Gbbb over C and a Pythagorean frequency ratio of *takes a deep breath* 4294967296/3486784401. I'd previously argued that the rank-2 way to analyze E half flat was as an Fb, i.e. a diminished fourth over C, with t(d4) =  8192/6561. Alsiadi goes one comma flatter than that. Good to know!

So start with a Pythagorean scale, and then flatten by two commas to get Alsiadi's microtones. And you can kind of decide for yourself whether that comma is the Pythagorean comma,
    t(A0) = 531441/524288

or the syntonic comma,
    t(Ac1) = 81/80

The first one is probably the historical origin of the microtones and the second one is a reinterpretation that keeps things spelled correctly and keeps the frequency ratios simpler.

Two stacked Pythagorean commas come to 46.9 cents, and two Syntonic commas come to 43 cents, and humans can only hear about 5 cents difference, so there isn't much that you can do to analyze actual performed music to decide between these as "the correct comma analysis".

Okay, I found some transcription errors on my part when doing the intervallic analysis of Alsiadi. I'm going to continue trying to find and fix those, and then we'll do a grand comparison of Alsiadi with maqamwolrd with the Turkish sources, and maybe with the ?oud.eclipse.co.uk? maqam bible PDF that might be the only source I really believe.

...

Okay, I've got all the obvious transcriptions errors cleaned up for maqamat starting on C, save for one. In Suzdilara (ascending) (Bayati ending), the B half flat is at 46 steps of 53-EDO (a rank-2 dddd10 over C or a rank-3 Acm7 over C) whereas it is at 47 steps of 53-EDO for the other two maqamat that start on C and have a B half flat (with 47 steps being a rank-2 dd9 over C or a rank-3 GrM7 over C), the other maqamat being Basandida (descending) and Nakriz (descending) and Rast (Ascending). I don't know what to do about this.

...

Woo! More progress. More transcription errors corrected. I've done a rank-2 analysis, and almost all of Alsiadi's pitches match things that I can derive from his commas. There are a few seemingly unresolvable differences though.

First up: "E-" is sometimes 15 steps of 53-EDO over C (i.e. a rank-2 dddd6, which is an Abbbbb over C) and sometimes it's 16 steps of 53-EDO (i.e. a rank-2 ddd5, which is a Gbbb over C). So it can be two or three commas flat relative to E natural.

The "B-" is also inconsistent: sometimes it's 46 steps of 53-EDO over C (i.e. a rank-2 dddd10, which is Ebbbb over C) and sometimes it's 47 steps over C (i.e. a rank-2 dd9 or Dbbb over C). Again, two or three commas flat (relative to B natural).

Now for some bigger problems: Suznak (ascending) is supposed to have an A half flat, apparently. But it happens at 36 steps over C, which is a rank-2 augmented fifth over C, better known as G#. If we follow in the footsteps of E- and B-, we'd expect it to be two or three steps flat relative to A natural, i.e. 37 or 38 steps over C (dddd9 or ddd8). MaqamWorld says that Suznak has an Ab, not an A-, but we don't have that either, we have a G# based on the commas. 

Maqam Suznak (ascending) is notated as ending with a Hijaz tetrachord. And looking into this, I found a different inconsistency in Alsiadi's writing: A hijaz tetrachord is usually [4, 14, 4] (in like 35 spots), but it's occasionally [5, 13, 4] (in about 5 spots). And actually, in Hijaz (descending), we get both varieties! 

    "Hijaz (descending)": [5, 13, 4, 9, 4, 14, 4], # Hijaz tetrachord + 9 + Hijaz tetrachord

Oh shit, it gets even worse. Here are the pitch classes for Hijaz (descending): 
    "Hijaz (descending)": "D, E-, F#, G, A, Bb, C#, D",

It's microtonal on the first tetrachord and it's tonal on the second?

Anyway, if you take Suznak (ascending) 
    "Suznak (ascending)": [9, 7, 6, 9, 5, 13, 4],

and replace the uncommon Hijaz with the common one,
    "Suznak (ascending)": [9, 7, 6, 9, 4, 14, 4],

then we get the pitch classes from Maqam world
        [C, D, E-, F, G, Ab, B, C]

with the Ab, but still no A-. If we suppose that A- is 2 or 3 commas below A natural, then we'd need a 6 or a 7 as the value for the fifth comma in the list, since A natural is 9 steps of 53-EDO above G natural. But I don't think [7, 11, 4] or [6, 12, 4] really deserve to be called a Hijaz tetrachord. I wan to change it to [4, 14, 4], but I'm feeling too lost at the moment. I'll figure it out soon enough though.

It shouldn't be this irregular. Pythagorean tuning and 53-EDO are very regular. All the major seconds are 9 steps, all the minor seconds are 4 steps, all the sharps are raised by 5 steps, all the flats are lowered by 5 steps. *with conviction* I'll figure it out soon enough though.

Next up, Hijaz ascending and descending are supposed to have E-, but they both occur just five commas above the tonic of D. However, 5 steps of 53-EDO is an augmented unison, so 5 steps over D is just a D#. This is four commas flat of an E natural, not two or three. Too flat, I say! It is a comma higher than an Eb though, so we're still in "the range between Eb and E natural" I guess.

Shad Araban (ascending) and (descending) only differ on the commas surrounding the "E" note. The staff notation says that they should both be Eb, but the commas say that it should be E- ascending and Eb descending. I'd like to clear this up by appealing to a different source, but I first looked at MaqamWorld and it's totally different. The MaqamWorld "Maqam Shadd ‘Araban" has a tonic on C instead of G, and it has no microtones (whereas Alsiadi's Shad has an A- regardless of the E-/Eb choice), and it doesn't have different ascending and descending versions. Alsiadi's ascending Shad and Maqam world's ascending Shadd are both notated with a Nakriz/Nikriz pentachord over a Hijaz tetrachord, but then the MaqamWorld version ascends over the octave with a Hijazkar pentachord that overlaps the Nikriz by three notes.

I also want to address issues with
* Bastah Nikar
* Farahnak (descending)
* Rahit Al Arwah (descending)

All of these three start on B-. And it seems that B- can be 46 or 47 steps of 53-EDo in Alsiadi's analysis. So when the pitches classes I derived for the maqamat using Alsiadi's commas (and a B- of 47 steps) didn't match the pitch classes he presented, I naturally tired a B- of 46 steps. It turns out, that doesn't solve any of the three and also causes lots of other maqamat to be misspelled.

With a B- at 47 steps, Farahnak (ascending) is spelled correctly but descending it looks like this:
    Farahnak (descending): [6, 9, 6, 7, 6, 9, 4]
        Alsiadi: [B-, C, D, E, F+, G, A, Bb]
        Derived: [B-, C, D, E-, F, E##, Dbbbbb, G##]

The steps that actually produce Alsiadi's pitch classes are:
    [6, 9, 9, 7, 6, 9, 4]

which only differs in turning a 6 into a 9 at the third place. I'm strongly inclined to call this a typo on his part.

Next, "Bastah Nikar". It's written the same ascending and descnding, which makes it a little less likely that my analysis is failing because of a typo on his part, but still possible.

With a B- at 47 steps, we get
Bastah Nikar: [7, 9, 7, 7, 4, 14, 4],
Alsiadi: [B-, C, D, E-, F, Gb, A, Bb]
Derived: [B-, B#, C##, Fb, Bbbbbb, E##, Dbbbb, B-]

The commas which produce alsiadi's pitch classes are: 
    [6, 9, 6, 7, 4, 14, 4]
or
    [6, 9, 7, 6, 4, 14, 4]

since there's a choice of E- (15 or 16 steps of 53-EDO).

If we use a B- at 46 steps, we get:
    Bastah Nikar: [7, 9, 7, 7, 4, 14, 4],
        Alsiadi: ['B-', 'C', 'D', 'E-', 'F', 'Gb', 'A', 'Bb']
        Derived: ['B-', 'C', 'D', 'E-', 'E#', 'F#', 'G##', 'A#']

and the commas which produce alsiadi's pitch classes are:
    [7, 9, 7, 6, 4, 14, 4],
or
    [7, 9, 6, 7, 4, 14, 4],

since again there's a choice of where to put E- (at 15 or 16 steps of 53-EDO).

So, .... those are all facts. Since D-natural is at 9 commas over C (a Pythagorean major second) and F natural is at 22 commas over C (a Pythagorean perfect fourth), the commas between them have to sum to 13 steps, not 14. I don't know how to fix this without appealing to other sources. We'll do that later.

Let's talk about Rahit Al Arwah (descending). This one is pretty obviously a typo on Alsiadi's part. He has written [6, 9, 4, 14, 6, 9, 4], but it should be [6, 9, 4, 14, 4, 9, 4]. The ascending form has the same pitch classes notated for all but the highest note (a Bb descending, a B- ascending), and also the same first two ajnas, so the two maqamat should should have the same lower steps/commas. Two lines of evidence that agree, so the third datum is probably in error. Also, between F# and A, the only options for a kind of G in 53-EDO or Pythagorean tuning are G natural or G sharp, which aren't found at 6 steps over F# (they're at 4 commas and 9 commas respectively), so the 6 steps thing isn't really an option, even if you ignore the ajnas and notated pitch classes and the ascending form.

Okay, that's it for "analyzing Alsiadi by comparing his pitch classes to his tetrachords". Time to get really really serious and analyze everything at once. One maqam at a time, all the sources, all the information, all the logic. We'll figure out the tetrachords too.

...

In fact, let's start with the tetrachords. Let's think of them as roughly Pythagorean. What should the neutral intervals be? Let's think of the neutral third to start. I hear Arabic music puts them rather close to the 24-EDO quarter tone, 350 cents, while Turkish music theory puts them closer to the justly tuned 5-limit rank-3 major third, t(M3) = 5/4 ~= 386 cents. For comparison, the 12-EDO major third is 400 cents and the Pythagorean major third is t(m3) = 81/64 ~= 408 cents. And for refresher, the just M3 is one justly tuned syntonic comma (t(Ac1) = 81/80 ~= 22 cents) lower than the Pythagorean one, and 53-EDO tunes both the syntonic comma and the Pythagorean comma (t(A0) = 531441/524288 ~= 23 cents) to 1 step, so we can pretty freely talk about lowering rank-2 intervals by A0 to get similar tuned frequencies as we would if we lowered them by a syntonic comma, Ac1.

If we lower the Pythagorean major third by one (Pythagorean) comma, we get a diminished fourth, d4, at 384 cents. Let's call that a Turkish neutral third in a rank-2 analysis. This was also a historically used value for analyzing the neutral third for Arabic music, e.g. the 
medieval Arabic music theorist Safi al-Din al-Urmawi outlined a  17-tone Pythagorean analysis of maqam pitches made by an extended spiral of perfect fifths, and the pitch between m3 and M3 was t(d4) = 8192/6561 ~=  384 cents.

Using this, we can give the steps between intervals of the Rast tetrachord as:
    [M2, d3, A1]

Or accumulative these to give intervals for each step relative to the tonic:
    [P1, M2, d4, P4]

If we lower the Pythagorean major third by two (Pythagorean) commas, we get a three-times diminished fifth, ddd5, tuned to 361 cents. Alsiadi uses intervals that are 2 or sometimes 3 commas flat from Pythagorean major to represent neutral intervals, so this is an option for Arabic rank-2 anslysis.

There are some other options we could consider for the representation of neutral thirds. Ancient music theorists thinking about arithmetic divisions of the neck of a string instrument (or, equivalently, harmonic means of frequency ratios) found that 
    (8/7) * (14/13) = 16/13 ~= 359 cents
    (8/7) * (13/12) = 26/21 ~= 370 cents

are easily constructed neutral thirds. The construction is basically starting with a string of length 16 units and dividing it in half, and then repeatedly dividing new segments in half:

    8:16 # 2/1
    8:12:16 # 3/2, 4/3
    8:12:14:16 # 3/2, 7/6, 8/7
    8:12:13:14:16 # 3/2, 13/12, 14/13, 8/7

And there's an interesting thing where even divisions of the string result in uneven divisions of the frequency ratios.  

I think the former neutral third, (8/7) * (14/13) = 16/13, is more natural, but the two are pretty similar. According to Margo Schulter, modern Syrian and historic Ottoman maqam music still use/used to use neutral thirds in this range.

I know I didn't explain that construction in detail, but perhaps you can trust me or infer how it is that by dividing a Pythagorean scale with harmonic means in a similar way, we can get a neutral third of
    (81/68) * (34/33) = 27/22 ~= 355 cents

And by taking certain harmonic divisions of the 5-limit m3 and M3 and combining them (or by just taking the mediant of 6/5 and 5/3), we can get a nice neutral third at
    (11/10) * (10/8) = 11/9 ~= 347 cents

So there are lots and lots of justifiable options available to use for a tuned neutral third if we skip the edifice of regular intervallic theory. Those all have factors of 11 or 13. If we just want to look at prime factors up of 7, then Ben Johnston's septimal super unison, t(Sp1) = 36/35 ~= 49 cents is a very good quarter tone, so adjusting the Pythagorean m3 and M3 by that gives us good neutral thirds of:
    
    t(SbGrm3) = 128/105 = (32/27) * (36/35) ~= 343 cents
    t(SpAcM3) = 315/256 = (81/64) / (36/35) ~= 359 cents

not that middle eastern theorists ever use these.

If Turkish and Arabic neutral thirds differ 30 cents or whatever - by more than a comma - I feel like I should be notating them differently/separately. But if they're both consistently using only one value for "neutral third", then it also feels like I should have a tuning-agnostic intervallic representation that shows how they're basically using the same scales, in so far as they are using the same scales. Idk what to do about cultural intonation.

Honestly, Alsiadi's two-comma-flat ddd5 at 361 cents as a neutral third is a nice compromise. It's pretty squarely between the modern arabic 24-EDO value of 350 cents and the maybe-Turkish 5-limit value at 386 cents. It's a modern fine-grained division produced by someone familiar with both Arabic and Turkish makams. One comma flat honestly doesn't sound very middle eastern; it's no more exotic than a five-limit just intonation scale, which music theory nerds frequently try to push as sounding more harmonious than 12-EDO. Two commas flat, in contrast, is clearly a different tone and makes it so that Rast is not just a major scale with a weird tuning, but it's own scale worth teaching. At 16 steps of 53-EDO, the two-commas flat neutral third also has a nice short name in 5-limit just intonation that can be spelled/enumerated correctly as a third: it's a Grave major third, GrM3, two syntonic commas below the Pythagorean M3, which is called an AcM3 in 5-limit just intonation. So two commas flat has a lot going for it.

That in mind, let's notate some ajnas using the idea that a half flat pitch class is two-commas flat from the natural version / the major Nth.

...

Archytas's Harmonic Means

There was a Greek music theorist named Archytas. He came up with some unusual musical scales. I haven't read translations of the primary sources, but my understanding is that he was enamored with super-particular ratios (those of the form {(n) / (n - 1)}) and the mathematical operation known as the harmonic mean.

The inverse of the harmonic mean of {n} numbers is the average of the inverses of the {n} numbers. Weird operation to define, right? For two numbers {a} and {b}, it has a simple form:

    H(a, b) = (2 * a * b) / (a + b)

Archytas, supposedly, would divide a frequency ratio {f} into smaller parts by taking the harmonic mean of {f} and {1/1}. The ratio of {f} to the harmonic mean of {f} and {1/1} was also of interest to him. Let's call it the complementary ratio. The complementary ratio is actually the arithmetic mean of {f} with 1/1. Anyway, let's call "harmonic mean with {f} and 1/1" the Archytas divisor of {f} to save space.

And now let's just do a few. The Archytas divisor of {2/1} is {4/3}:

    H((2/1), (1/1)) = (4/3)

The complementary (arithmetic) ratio is {3/2} because

    (2/1) / (4/3) = (3/2)

or because

    ((2/1) + (1/1)) / 2 = (3/2)

The Archytas divisor of (4/3) is (8/7) with a complement of (7/6).

The Archytas divisor of (3/2) is (6/5) with a complement of (5/4).

You can see that these are all super-particular ratios. Also, for a super particular ratio with numerator {n}, the Archytas divisor will have numerator {2 * n} and the denominator will be {2 * n - 1}. Further, the complement of the divisor will be {n - 1 / n - 2}.

A little algebra will show that the product of the Archytas divisor and its complement reproduces the original ratio:

    ((2n) / (2n - 1)) * ((2n - 1) / (2n - 2)) = n / (n - 1)

.

Here's a table of some Archytas divisors:

Arch(2/1) = 4/3 & 3/2
Arch(4/3) = 8/7 & 7/6
Arch(8/7) = 16/15 & 15/14
Arch(16/15) = 32/31 & 31/30
Arch(15/14) = 30/29 & 29/28
Arch(7/6) = 14/13 & 13/12
Arch(14/13) = 28/27 & 27/26
Arch(13/12 v (1/1) = 26/25 & 25/24
Arch(3/2) = 6/5 & 5/4
Arch(6/5) = 12/11 & 11/10
Arch(12/11) = 24/23 & 23/22
Arch(11/10) = 22/21 & 21/20
Arch(5/4) = 10/9 & 9/8
Arch(10/9) = 20/19 & 19/18
Arch(9/8) = 18/17 & 17/16
Arch(18/17) = 36/35 & 35/34
Arch(17/16) = 34/33 & 33/32
.

I don't really know why, but Archytas liked the ratio (28/27) and used it in his scales a lot. I think it's interesting to imagine a world where Archytas had more influence and we used septimal frequency ratios like (8/7) and (7/6) and (28/27) everywhere in our music.

Here's a thought experiment: if we think of 8/7 and 7/6 as halves of 4/3, perhaps a small half and a large half, then what ratios are the one-quarters and three-quarters of 4/3?

The smaller half, 8/7, divides into 16/15 and 15/14. The large half, 7/6, divides into 14/13 and 13/12. Those are the four Archytas quarters of 4/3.

If we try to recombine the four Archytas quarters, we get these as our derived halves:

(16/15) * (15/14) = 8/7 # 231 cents

(16/15) * (14/13) = 224/195 # 240 cents

(16/15) * (13/12) = 52/45 # 250 cents

(15/14) * (14/13) = 15/13 # 248 cents

(15/14) * (13/12) = 65/56 # 258 cents

(14/13) * (13/12) = 7/6 # 267 cents

Since 4/3 is 498 cents, the actual half of it is 249 cents, which is really close to both 15/13 and 52/45. These two also produce 4/3 exactly when multiplied. So they're also very good halves of 4/3.

The first of these, 15/13, is called the justly tuned value for a "recessed acute augmented second", ReAcA2, in my rank-6 Lilley-Johnston system for naming intervals, because it's one tridecimal comma smaller than the acute augmented second:

    t(Pr1) = (65/64)

    t(AcA2) = (75/64)

    AcA2 - Pr1 = ReAcA2

   (75/64) / (65/64)  = (15/13)

and prominent pairs with recessed for describing rank-6 intervals, such as those that are justly tuned to tridecimal frequency ratios.

The second one, 52/45, is a a justly tuned "prominent grave diminished third", PrGrd3, since a grave diminished third is justly tuned to (256/225).

If we multiply the Archytas halves (8/7, 7/6) of 4/3 by the quarters (16/15, 15/14, 14/13, 13/12), we get a bunch of three-quarters:

     (8/7 * 16/15) = (128/105) # 343 cents

(8/7 * 15/14) = (60/49) # 351 cents 

(7/6 * 14/13) = (49/39) # 395 cents

(7/6 * 13/12) = (91/72) # 405 cents

(7/6 * 16/15) = (56/45) # 379 cents

(7/6 * 15/14) = (5/4) # 386 cents

(8/7 * 14/13) = (16/13) # 360 cents

(8/7 * 13/12) = (26/21) # 370 cents

but only the last four of these can be gotten by dividing (4/3) by the quarters:

     (4/3) / (16/15) = 5/4 # 386 cents

(4/3) / (15/14) = 56/45 # 379 cents

(4/3) / (14/13) = 26/21 # 370 cents

(4/3) / (13/12) = 16/13 # 360 cents

so I think of them as the truer three-quarters of 4/3 perhaps. The weird ones can probably be used to get weird flavors of not-quite 4/3, if that's something you might like.

If we take Archytas means and complements of the eighths of 4/3, namely (13/12, 14/13, 15/14, 16/15), then we get super particular ratios from (25/24, 26/25, 27/26, ... all the way up to ..., 31/30, 32/31). 

I don't know that I want to figure out cents for all of them, but the middle ones multiplied together give

    (28/27) * (29/28) = 29/27

which we might think of as a very good non-Archytas fourth of 4/3, just as 15/13 and 52/45 were very good non-Archytas halves.

If we try breaking 15/13 and 52/45 into Archytas means and complements, we get:

    15/14 # 119 cents  // Archytas mean of 15/13

    104/97 # 121 cents // Archytas mean of 52/45

    14/13 # 128 cents // Archytas complement of 15/13

    97/90 # 130 cents // Archytas complement of 52/45

If we take the middle two and recombine them, we get even better approximations for the actual frequency-space half of (4/3), and the combination is 112/97.

I wonder if any of these are mathematically optimal rational halves for 4/3 in any sense, like truncations of the continued fraction expansion of sqrt(4/3). The continued fraction is [1, 6, 2, 6, 2, 6, 2, ...]. 

When I did that calculation, I messed up and got sad and tried something else. Since 

    sqrt(4/3) = 2/3 * sqrt(3)

I just looked up truncations of the C.F. for sqrt(3) and multiplied them by 2/3. Possibly mathematically equivalent? I don't know. But it worked. These guys approximate sqrt(3): 

    7/4, 26/15, 97/56, ...

and scaled by 2/3 we get

    7/6, 52/45, 97/84, ...

And the first two of those have already made an appearance. Which just goes to show that sometimes when you calculate rational approximations to sqrt(4/3) by one method, you get some overlap with approximations to sqrt(4/3) produced by a second method.

Since 97 is quite a large prime as far as audible harmonics are concerned, I don't have much use for it in extended just intonation. And the next term in the series has a prime factor of 607, which is even worse. But I'm glad we looked all the same.

...

Tempered Frequency Ratios? Why, I never!

On the Xenharmonic wiki, people constantly talk about tempered frequency ratios. Tempering an interval is a normal operation: it means tuning an interval, like a diminished second, to a frequency ratio of 1/1.

    t(d2) = 1/1

Tuning systems perform that mapping: from intervals to frequency ratios.

"Tempering a frequency ratio" is a type error. You can't tune 5/3 to 1/1. Five thirds does not equal one.

This frustrated me for a long time, because many of the Xenharmonic people know more math than me and sometimes when I think I've come up with a new result in music theory, I'll find it on their wiki and feel sad that I got scooped by some crayon-eating group theory savant who thinks 5/3 = 1/1.

I'm not being too harsh. Look at this shit: "Tempering [441/440] out splits 11/10 into two even halves [and] equates 21/16 with 55/42."

I think I figured out some of the valid math that they're describing in an invalid way today, and it's not all that complex.

So, let's find "positive 7-limit frequency ratios that are tempered out by 53-EDO". We'll start by finding the approximate number of steps of 53-EDO we need to reach different prime harmonics up to the one with frequency ratio 7/1.

If

    2^(i / 53) = (2/1)

then

    i for 2/1 = 53 * log(2/1) / log(2)) = 53

We don't even have to round that one. But in general, we want an integer number of 53-EDO steps, so we will round that middle expression to the nearest integer. Here's the formula for the other primes:

    i for 3/1 = round(53 * log(3/1) / log(2)) = 84

    i for 5/1 = round(53 * log(5/1) / log(2)) = 123

    i for 7/1 = round(53 * log(7/1) / log(2)) = 149

It doesn't matter so far if you know the name of the interval that is justly tuned to 7/1. Whatever the interval name is, you know how many steps it will be in 53-EDO. Or rather, we're defining a 53-EDO in which that interval is tuned to 149 steps. And it will be good enough, because 2^(149/53) = 7.01926881203, which is basically 7; don't quibble.

We've basically just defined a tuning system without knowing the basis intervals. We can use this half-assed monstrosity to find the 53-EDO frequency ratios for other intervals we don't know the names  of: we just have to know how to build them from the prime basis intervals.

So, for example, there's a famous interval made from (whatever interval is justly tuned to 3/1) minus (whatever interval is justly tuned to 2/1). It happens to be the perfect fifth, but that's currently irrelevant. To tune this mystery difference intervals, we just do arithmetic with our EDO steps: 84 - 53 = 31. And, look, 

    2^(31/53) = 1.49994090308

is a fine approximation for the traditional just tuning of (the mystery difference interval), which is 

    3/2 = 1.5.

To find tempered out intervals, we just find coordinates (a, b, c, d) such that {steps = 0} in

    steps = a * 53 + b * 84 + c * 123 + d * 149

Now you can do a search over coordinates, with {a} from, oh, -10 to 10, and {b} from whatever to whatever, and so on. Find any set of four coordinates that will give 0 steps. Then find the just frequency ratios for each of those sets of coordinates.

The actual frequency ratio for a given number of steps {i} in our 53-EDO is

    2^(i/53)

but the just frequency ratio that we're approximating with 53-EDO is:

    frequency_ratio = (2/1)^a * (3/1)^b * (5/1)^c * (7/1)^d

Do that for every tempered set of coordinates (a, b, c, d), and you'll get a list of justly tuned frequency ratios for unknown, unnamed intervals that are tempered out by 53-EDO:

1.0 = 1/1
1.0002286236854139 = 4375/4374
1.001129150390625 = 32805/32768
1.0013580322265625 = 65625/65536
1.0031020408163265 = 6144/6125
1.003331373701744 = 5120/5103
1.004234693877551 = 19683/19600
1.0044642857142858 = 225/224
1.0046939300411524 = 15625/15552
1.0075801749271136 = 1728/1715
1.0078105316200554 = 4000/3969
1.0087178844752187 = 177147/175616
1.0089485012755102 = 50625/50176
1.0091791708002646 = 390625/387072
1.0109368010242645 = 65536/64827
1.0120783007080383 = 2430/2401
1.0123096857790734 = 3125/3087
1.0154499117431228 = 51200/50421
1.0165965074076277 = 273375/268912
1.01682892544773 = 78125/76832
1.0197500312673413 = 839808/823543
1.0199831702776905 = 120000/117649
1.024536666573573 = 843750/823543

This list has famous fractions that people made stupid folksy names for because they didn't have a regular system for naming intervals. Like 4375/4374 is "ragisma". And 32805/32768 is "schisma". And 65625/65536 is "The Horwell comma". And 6144/6125 is "the porwell comma". And 15625/15552 is "kleisma" and it just goes on and on. Now you've got a wiki full of people saying shit like "sabric pote porkypine7 eris wa, hyper-Partchian monzo mos wedgie, unidecimal marvel[22] hobbit, keemic fourthward alpharabian sevond, oneirotonic semihard tamnams, zg32 zotrigu comma".

I don't really hate these people. I know that I'm basically one of them. But I want them to use real words. Weird music, standard math, real words; I swear, it's a good combination.

In summary, you can define an EDO tuning system by finding a number of EDO steps that approximates each prime harmonic up to a desired limit, and then use the intervals that are justly associated with the prime harmonics as an unnamed, unspoken basis for all of your intervals, and then you can even find new intervals that are tempered out by the half-constructed EDO tuning system, and give their just frequency ratios new horrible unsystematic names. It's fun! It's also pretty fast. Want to see 11-limit just frequency ratios for some unnamed intervals that are tempered out by, oh, 31-EDO?

The steps for the 11-limit basis intervals are found like this:

round(31 * log(2/1) / log(2)) = 31 
round(31 * log(3/1) / log(2)) = 49
round(31 * log(5/1) / log(2)) = 72
round(31 * log(7/1) / log(2)) = 87
round(31 * log(11/1) / log(2)) = 107

So the number of 31-EDO steps for an arbitrary interval in the 11-limit prime basis will be

    steps = a * 31 +  b * 49 + c * 72 + d * 87 + e * 107

Here are 11-limit justly tuned frequency ratios for some rank-5 intervals that are tempered out by the 31-EDO that we constructed from the harmonics, along with coordinates in parentheses from the prime harmonic interval basis:

1/1 (0, 0, 0, 0, 0)
3025/3024 (-4, -3, 2, -1, 2)
2401/2400 (-5, -1, -2, 4, 0)
540/539 (2, 3, 1, -2, -1)
1375/1372 (-2, 0, 3, -3, 1)
441/440 (-3, 2, -1, 2, -1)
385/384 (-7, -1, 1, 1, 1)
3136/3125 (6, 0, -5, 2, 0)
3388/3375 (2, -3, -3, 1, 2)
243/242 (-1, 5, 0, 0, -2)
225/224 (-5, 2, 2, -1, 0)
6912/6875 (8, 3, -4, 0, -1)
176/175 (4, 0, -2, -1, 1)
1331/1323 (0, -3, 0, -2, 3)
3773/3750 (-1, -1, -4, 3, 1)
2835/2816 (-8, 4, 1, 1, -1)
1728/1715 (6, 3, -1, -3, 0)
2420/2401 (2, 0, 1, -4, 2)
126/125 (1, 2, -3, 1, 0)
121/120 (-3, -1, -1, 0, 2)
1944/1925 (3, 5, -2, -1, -1)
99/98 (-1, 2, 0, -2, 1)
1617/1600 (-6, 1, -2, 2, 1)
2430/2401 (1, 5, 1, -4, 0)
81/80 (-4, 4, -1, 0, 0)
1815/1792 (-8, 1, 1, -1, 2)
3168/3125 (5, 2, -5, 0, 1)
2662/2625 (1, -1, -3, -1, 3)
2187/2156 (-2, 7, 0, -2, -1)
8712/8575 (3, 2, -2, -3, 2)
2541/2500 (-2, 1, -4, 1, 2)
891/875 (0, 4, -3, -1, 1)
3993/3920 (-4, 1, -1, -2, 3)
8748/8575 (2, 7, -2, -3, 0)
9801/9604 (-2, 4, 0, -4, 2)
5103/5000 (-3, 6, -4, 1, 0)
3267/3200 (-7, 3, -2, 0, 2)
8019/7840 (-5, 6, -1, -2, 1)
6561/6400 (-8, 8, -2, 0, 0)

The 3025/3024 ratio is called the "lehmerisma". The 2401/2400 is "breedsma". The 540/539 is "swetisma". The 441/440 is... "Werckmeister's undecimal septenarian schisma". 3136/3125 is the "hemimean comma". A lot of these are superparticular ratios. They're kind of pretty if you don't look up the Xenharmonic names.

Of the commas, I thought that these four were not constructible from sums of the others:

[-4, -3, 2, -1, 2]: 3025/3024 [-5, -1, -2, 4, 0]: 2401/2400 [2, 3, 1, -2, -1]: 540/539 [6, 0, -5, 2, 0]: 3136/3125

and that sums of those four could be used to make all of the others. But when I do a larger search over value ranges for the components, (a, b, c, d, e), my procedure for generating that smell summary set gives different values, so maybe it's nonsense. It's not an intrinsic fact of the EDO.

I did a search for a fifth vector that can be combined with those four to make a unimodular matrix, and I found all of these:

[-5, 1, 0, 0, 1] # 33/32
[2, 2, -1, -1, 0] # 36/35
[-2, 2, 1, 0, -1] # 45/44
[-4, -1, 0, 2, 0] # 49/48
[1, 0, 2, -2, 0] # 50/49
[-1, -3, 1, 0, 1] # 55/54
[3, 0, -1, 1, -1] # 56/55
[0, -1, -2, 1, 1] # 77/75
[2, -2, 2, 0, -1] # 100/99
[0, -5, 1, 2, 0] # 245/243
[5, 0, 0, -3, 1] # 352/343
[3, 4, -4, 0, 0] # 648/625

And many more complex ones. These are all tuned to one step of 31-EDO.

I did the same procedure for rank-5 coordinates with 19-EDO. These are "the four independent commas" which are probably just noise induced by the extent of my search of parameter space:

[6, 0, -5, 2, 0] = 3136/3125
[-1, -7, 4, 1, 0] = 4375/4374
[2, 3, 1, -2, -1] = 540/539
[-6, 6, -3, 0, 1] = 8019/8000

I searched for a coordinates of a fifth interval that would make the total matrix unimodular, and found all of these:

[-2, 1, -1, 1, 0] # 21/20
[-3, -1, 2, 0, 0] # 25/24
[2, -3, 0, 1, 0] # 28/27
[-5, 1, 0, 0, 1] # 33/32
[0, -1, 1, 1, -1] # 35/33
[2, 2, -1, -1, 0] # 36/35
[1, 0, 2, -2, 0] # 50/49
[-1, -3, 1, 0, 1] # 55/54
[0, -1, -2, 1, 1] # 77/75
[4, 0, 1, -1, -1] # 80/77
[0, 4, 0, -1, -1] # 81/77
[-1, 2, 0, -2, 1] # 99/98
[4, 0, -2, -1, 1] # 176/175
[-4, 3, 0, 1, -1] # 189/176

And many more complex ones. These are all coordinates for rank-5 intervals in the-basis-of-unnamed-prime-harmonic-intervals that are tuned to one step of 19-EDO in 19-EDO.

Here are some intervals tempered out by 53-EDO:

[-1, 2, 0, -2, 1] # 99/98 [-3, -1, -1, 0, 2] # 121/120 [4, 0, -2, -1, 1] # 176/175 [-5, 2, 2, -1, 0] # 225/224 [-7, -1, 1, 1, 1] # 385/384 [2, 3, 1, -2, -1] # 540/539


And here are coordinates for some simple intervals that are tuned to one step of 53-EDO along with their just tunings:

[1, 0, 2, -2, 0] # 50/49
[-1, -3, 1, 0, 1] # 55/54
[6, -2, 0, -1, 0] # 64/63
[-4, 4, -1, 0, 0] # 81/80
[2, -2, 2, 0, -1] # 100/99
[1, 2, -3, 1, 0] # 126/125
[-1, 5, 0, 0, -2] # 243/242
[0, -5, 1, 2, 0] # 245/243
[5, 0, 0, -3, 1] # 352/343

.

Cool.

Rationalizing 53-EDO For Turkish Maqam Analysis

I've heard that 53-EDO is used productively to analyze Turkish music. It can be defined by pure octaves and tempering out ddddddd6, which as coordinates (0, -1) in the (A1, d2) basis.

    t(P8) = 2/1

    t(ddddddd6) = 1/1

The step size of 53-EDO, 2^(1/53), is 

    1200 * log(2^(1/53)) / log(2) = 1200/53 ~ 22.6 cents

which is really close to the Pythagorean comma, used in analyzing rank-2, 3-limit just intonation music. The Pythagorean comma is the augmented zeroth interval, A0, which Pythagorean tuning tunes to a frequency ratio of 531441/524288 ~= 23.46 cents. Quite close to the step of 53-EDO at 22.6 cents.

The step of 53-EDO is also quite close to the justly tuned syntonic comma, also called the acute unison or Ac1. In normal 5-limit just intonation, this interval is tuned to a frequency ratio of 81/80 ~= 21.5 cents. The acute unison is used in analyzing rank-3, 5-limit just intonation music. Humans can only discriminate about 5-cent differences in frequency, so the justly tuned A0, the 53-EDO tuned A0, and the 5-limit Ac1 are are basically equal as far as we can hear.

I haven't read original sources from Turkish music theorists on 53-EDO, but just the fact that the two commas are mentioned as justifications for 53-EDO in western sources makes me want to give 53-EDO both a rank-2, 3-limit and a rank-3, 5-limit interpretation.

Let's start by finding rank-1 intervals with short names for every 53-EDO step. To do this, I just tune various small intervals, increasing the complexity until I've got one tuned to each of the 53 steps. The results are:

2^(0/53) - P1 : (0, 0)
2^(1/53) - A0 : (0, -1)
2^(2/53) - dddd4 : (1, 3)
2^(3/53) - dd3 : (1, 2)
2^(4/53) - m2 : (1, 1)
2^(5/53) - A1 : (1, 0)
2^(6/53) - AA0 : (1, -1)
2^(7/53) - ddd4 : (2, 3)
2^(8/53) - d3 : (2, 2)
2^(9/53) - M2 : (2, 1)
2^(10/53) - AA1 : (2, 0)
2^(11/53) - dddd5 : (3, 4)
2^(12/53) - dd4 : (3, 3)
2^(13/53) - m3 : (3, 2)
2^(14/53) - A2 : (3, 1)
2^(15/53) - dddd6 : (4, 5)
2^(16/53) - ddd5 : (4, 4)
2^(17/53) - d4 : (4, 3)
2^(18/53) - M3 : (4, 2)
2^(19/53) - AA2 : (4, 1)
2^(20/53) - ddd6 : (5, 5)
2^(21/53) - dd5 : (5, 4)
2^(22/53) - P4 : (5, 3)
2^(23/53) - A3 : (5, 2)
2^(24/53) - dddd7 : (6, 6)
2^(25/53) - dd6 : (6, 5)
2^(26/53) - d5 : (6, 4)
2^(27/53) - A4 : (6, 3)
2^(28/53) - AA3 : (6, 2)
2^(29/53) - ddd7 : (7, 6)
2^(30/53) - d6 : (7, 5)
2^(31/53) - P5 : (7, 4)
2^(32/53) - AA4 : (7, 3)
2^(33/53) - dddd8 : (8, 7)
2^(34/53) - dd7 : (8, 6)
2^(35/53) - m6 : (8, 5)
2^(36/53) - A5 : (8, 4)
2^(37/53) - dddd9 : (9, 8)
2^(38/53) - ddd8 : (9, 7)
2^(39/53) - d7 : (9, 6)
2^(40/53) - M6 : (9, 5)
2^(41/53) - AA5 : (9, 4)
2^(42/53) - ddd9 : (10, 8)
2^(43/53) - dd8 : (10, 7)
2^(44/53) - m7 : (10, 6)
2^(45/53) - A6 : (10, 5)
2^(46/53) - dddd10 : (11, 9)
2^(47/53) - dd9 : (11, 8)
2^(48/53) - d8 : (11, 7)
2^(49/53) - M7 : (11, 6)
2^(50/53) - AA6 : (11, 5)
2^(51/53) - ddd10 : (12, 9)
2^(52/53) - d9 : (12, 8)
2^(53/53) - P8 : (12, 7)

I've also listed the interval coordinates in Lilley's (A1, d2) basis at the end of each line.

Supposedly not all of these steps are used in Turkish music. Unless you're transposing around maybe. But normal practice is to just use intervals with step sized of: [0, 4, 5, 8, 9, 13, 14, 17, 18, 22, 23, 26, 27, 30, 31, 35, 36, 39, 40, 44, 45, 48, 49]. This set makes a lot of sense! If we just remove all the lines that have highly modified intervals - those that are twice or more diminished or twice or more augmented -, then we get the Turkish set at steps [0, 4, 5, ...] plus 

2^(1/53) - A0 : (0, -1)

2^(52/53) - d9 : (12, 8)

which are just the 53-EDO step and its octave complement. A0 is also the formal name for the Pythagorean comma. 

Here are the rest of the natural and once-modified intervals, the ones corresponding to steps of [0, 4, 5, 8, 9, 13, 14, 17, 18, 22, 23, 26, 27, 30, 31, 35, 36, 39, 40, 44, 45, 48, 49, 52]:

2^(0/53) - P1 : (0, 0)
2^(4/53) - m2 : (1, 1)
2^(5/53) - A1 : (1, 0)
2^(8/53) - d3 : (2, 2)
2^(9/53) - M2 : (2, 1)
2^(13/53) - m3 : (3, 2)
2^(14/53) - A2 : (3, 1)
2^(17/53) - d4 : (4, 3)
2^(18/53) - M3 : (4, 2)
2^(22/53) - P4 : (5, 3)
2^(23/53) - A3 : (5, 2)
2^(26/53) - d5 : (6, 4)
2^(27/53) - A4 : (6, 3)
2^(30/53) - d6 : (7, 5)
2^(31/53) - P5 : (7, 4)
2^(35/53) - m6 : (8, 5)
2^(36/53) - A5 : (8, 4)
2^(39/53) - d7 : (9, 6)
2^(40/53) - M6 : (9, 5)
2^(44/53) - m7 : (10, 6)
2^(45/53) - A6 : (10, 5)
2^(48/53) - d8 : (11, 7)
2^(49/53) - M7 : (11, 6)

Those are normal things! We know how to use them fairly well in western music theory! We probably don't even need a rank-3 analysis with the syntonic comma, but I intend to do that anyway.

This also has intervals between the minor and major versions of (2nds, 3rds, 6ths, 7ths) which is good. Middle eastern music definitely has extra/intermediate/neutral versions of the imperfect ordinal intervals.

If you look up the names for the notes associated with the Turkish subset of 53-EDO on Wikipedia, you get a table with a bunch of Turkish words you can't read and some accidentals you've never seen before (in the Arel-Ezgi-Uzdilek notation column). The analysis above does away with that! We don't need to call 2^(5/53) "the thing that's one comma sharp of a minor second" with a new symbol, it's just an augmented unison! This analysis also reveals what I'm tempted to call a mistake in the Arel-Ezgi-Uzdilek system: F# and Gb aren't the same in 53-EDO: 26 steps is the tuned frequency for the diminished 5th, i.e. Gb over C, and 27 steps is tuned frequency for the augmented 4th, i.e. F# over C. But A-E-U have them both at 26 steps.

I'm going to want to refer to that list of intervals frequently, so I'll condense it here: these are the intervals of Turkish music we arrive at from the rank-2 analysis of 53-EDO with restrictions to steps [0, 4, 5, ...] and so on: 

    [P1, m2, A1, d3, M2, m3, A2, d4, M3, P4, A3, d5, A4, d6, P5, m6, A5, d7, M6, m7, A6, d8, M7]

These natural and once-modified intervals appear in the same order in 53-EDO as they do in Pythagorean tuning, but in a different order from quarter comma meantone or 5-limit just intonation.

How do we make a rank-3 version of 53-EDO?

I don't really know. I haven't thought a ton about turning 5-limit just intonation into different EDOs. But I think I've got a solution for this specific EDO. Obviously we still want pure octaves. And it would be ideal if the acute unison, Ac1, was tuned to 1^(1/53). That's two parameters fixed. If they're independent, then we still need to fix a third parameter. I think they are independent, and I think the third parameter is naturally fixed by tuning the interval Gr5, which has coordinates (1, 7, 4) in the rank-3 Lilley basis of (Ac1, A1, d2). Normally it's tuned to 40/27 in 5-limit just intonation. We could also use the interval with coordinates (3, 5, 3), which 5-limit just intonation tunes to 2187/1600. Not as good. The (1, 7, 4) interval makes the system of basis vectors of have determinant -1 and the (3, 5, 3) intervals creates a basis matrix with determinant +1. They both make unimodular systems.

Next I want a function to convert interval coordinates from the rank-3 Lilley basis, (Ac1, A1, d2) to this new basis, (Ac1, Gr5, P8). If you've been reading this blog for a while, you probably know the drill: For a change of basis, find the basis vectors of the old system expressed in the new system:

(1, 0, 0) = Ac1
(5, 7, -4) = A1
(-9, -12, 7) = d2

And convert rows to columns:

def convert_rank3_Lilly_basis_to_53edo_basis(interval):
(x, y, z) = interval
a = x * 1 + y * 5 + z * -9
b = x * 0 + y * 7 + z * -12
c = x * 0 + y * -4 + z * 7
return (a, b, c)

From the old coordinates (in the rank-3 Lilley basis for 5-limit just intonation intervals) like  

d1 = (0, -1, 0)
P1 = (0, 0, 0)
d2 = (0, 0, 1)
A1 = (0, 1, 0)
m2 = (0, 1, 1)
M2 = (0, 2, 1)
A2 = (0, 3, 1)
d3 = (1, 2, 2)
m3 = (1, 3, 2)
M3 = (1, 4, 2)
d4 = (1, 4, 3)
A3 = (1, 5, 2)
P4 = (1, 5, 3)
A4 = (1, 6, 3)
d5 = (2, 6, 4)
P5 = (2, 7, 4)
d6 = (2, 7, 5)
A5 = (2, 8, 4)
m6 = (2, 8, 5)
M6 = (2, 9, 5)
A6 = (2, 10, 5)
d7 = (3, 9, 6)
m7 = (3, 10, 6)
M7 = (3, 11, 6)
d8 = (3, 11, 7)
A7 = (3, 12, 6)
P8 = (3, 12, 7)

and this change of basis function, we can now get interval coordinates in the (Ac1, Gr5, P8) basis, which we can then tune using the constraints

    t(Ac1) = 2^(1/53)

    t(Gr5) = ???

    t(P8) = 2/1

to investigate whether we can get a 53-EDO with a rank-3 intervallic interpretation, so that we can reverse the interpretation and rational 5-limit just intonation to analyze 53-EDO Turkish music. It's a good plan.

I tried t(Gr5) = 40/27, and it looks bad. And of course it does; there's no place for nice 5-limit fractions in 53-EDO. Let's now try the closest number of steps in 53-EDO to 40/27. We can find this by inverting

    2^(x / 53) = 40/27

    x = 53 * log(40/27) / log(2) = 30.053...

So let's try t(Gr5) = 2^(30/53). Hopefully once we do this, we'll get a a 53-EDO, and from the intervals that are tempered out, we can find a better way to define 53-EDO.

Woo! It's working really well. There were lots of intervals tuned to the same frequency ratios. The difference between any two intervals tuned to the same frequency ratio will be another interval which the tuning system tempers out. There happened to be two intervals that showed up again and again. I'll just show one example of each. The acute unison and the grave diminished second were tuned to the same 53-EDO frequency ratio. From this, we can say that the difference between them is tempered out. I'll show the difference using coordinates in the rank-3 Lilley basis:

    Ac1 - Grd2 :: [1, 0, 0] - [-1, 0, 1] = [2, 0, -1]
    Grd2 - Ac1 :: [-1, 0, 1] - [1, 0, 0] = [-2, 0, 1]

The difference in either direction is tempered out, so I've shown  both differences. You just negate the coordinates.

The second tempered out interval can be see as the difference between a grave acute unison and a diminished second:

    GrA1 - d2 :: [-1, 1, 0] - [0, 0, 1] = [-1, 1, -1]
    d2 - GrA1 :: [0, 0, 1] - [-1, 1, 0] = [1, -1, 1]

Are you shaking with excitement yet? These are both forms of famous intervals! The interval [2, 0, -1] is the grave grave diminished zeroth, GrGrd0, better known as the Schisma! In 5-limit just intonation, it's tuned to 32805/32768. The interval [-1, 1, -1] is an Acute diminished diminished zeroth, Acdd0, better known as the Kleisma! In 5-limit just intonation, it's tuned to 15625/15552. So it looks like 53-EDO might be definable by pure octaves and tempering out the Schisma and the Kleisma?

Sadly, this fact is stated fairly openly on the wikipedia page for 53-EDO:

"The 53-TET tuning equates to the unison, or tempers out, the intervals 32805/32768, known as the schisma, and 15625/15552, known as the kleisma. ... The fact that 53 ET tempers out both characterizes it completely as a 5 limit temperament: it is the only regular temperament tempering out both of these intervals, or commas."

But I never knew what it meant before! Because it's mathematical nonsense to equate a positive fraction like 32805/32768 with unison. I still stand by that! The wikipedia page is obviously and pathetically wrong. And it's so sad. Someone clearly did the math right, but then someone wrote about it very wrong. Maybe even the same person. It turns out that the Japanese guy who named the Kleisma also figured out that 53-EDO tempers it out, and he popularized it, and he very well might be the reason why modern Turkish music theorists think 53-EDO is a cool thing to use for analysis. Sorry, Shohe Tanaka. I hope you knew the difference between intervals and frequency ratios. If so, we've done you dirty. If not, I'm still impressed that you got the math right, if not the communication of it.

Okay, rant over and nods to history completed. Now what are the actual rank-3 interval names for the steps of 53-EDO? How are those intervals justly tuned? How do the 5-limit just tunings of the rank-3 intervals compare to the 3-limit/Pythagorean tunings of the rank-2 intervals?

Here are some short rank-3 intervals for each 53-EDO step:

0: P1
1: Ac1, Grd2
2: GrA1, d2
3: A1, Acd2
4: AcA1, Grm2
5: m2
6: Acm2
7: GrM2
8: M2
9: AcM2
10: GrA2, Grd3
11: A2, d3
12: AcA2, Acd3
13: Grm3
14: m3
15: Acm3
16: GrM3
17: M3
18: AcM3, Grd4
19: GrA3, d4
20: A3, Acd4
21: AcA3, Gr4
22: P4
23: Ac4
24: GrA4
25: A4
26: AcA4
27: Grd5
28: d5
29: Acd5
30: Gr5
31: P5
32: Ac5, Grd6
33: GrA5, d6
34: A5, Acd6
35: AcA5, Grm6
36: m6
37: Acm6
38: GrM6
39: M6
40: AcM6
41: GrA6, Grd7
42: A6, d7
43: AcA6, Acd7
44: Grm7
45: m7
46: Acm7
47: GrM7
48: M7
49: AcM7, Grd8
50: GrA7, d8
51: A7, Acd8
52: AcA7, Gr8
53: P8

If we restrict ourselves to steps [0, 4, 5, 8, 9, 13, 14, 17, 18, 22, 23, 26, 27, 30, 31, 35, 36, 39, 40, 44, 45, 48, 49], we get... this set:

0: P1
4: AcA1, Grm2
5: m2
8: M2
9: AcM2
13: Grm3
14: m3
17: M3
18: AcM3, Grd4
22: P4
23: Ac4
26: AcA4
27: Grd5
30: Gr5
31: P5
35: AcA5, Grm6
36: m6
39: M6
40: AcM6
44: Grm7
45: m7
48: M7
49: AcM7, Grd8

Which is okay, I guess? It has the perfect, minor, and major intervals. Then it also has all of [Ac, AcM, AcA, Gr, Grm, Grd] as qualities, which are a little weird, but it's not like we were going to avoid them when we defined the system with Ac1 at step 1 and Gr5 at step 30. This 3-limit interval interpretation of 53-EDO also doesn't have neutral seconds, thirds, sixths, or sevenths, which doesn't look good to me.

Of the rank-2 and rank-3 intervals that have the same names and have that name appearing in both the rank-2 and rank-3 analyses of 53-EDO, there has been some movement. The Pythagorean m2 was at 4 steps, while the 5-limit one is at 5-steps. The Pythagorean M2 narrowed from 9 steps to 8 steps. The m3 widened from 13 to 14 steps. The M3 narrowed from 18 to 17 steps. The perfect intervals all stayed put, and the motions of the 6ths and 7ths can be inferred from the seconds and thirds by octave complementation.

It's kind of interesting that e.g. Turkish music using both steps 4 and 5 means that you have a minor second available to you in either analysis system. The consecutively used steps appear in the right places.

Okay, I'm going to be real with you: I don't know how to use Acute and Grave intervals. "Acute" and "Grave" only exist as modifiers so that we still have names for the 3-limit Pythagorean intervals when we make the imperfect intervals become 5-limit. A tuning system where a quarter of the notes have names like "AcA5" isn't actually that useful to me. The rank-2 analysis is way more compelling. And this saddens me, because Shohe Tanaka thought that 53-EDO was good for representing 5-limit just intonation, and I'm not seeing it.

I will admit that, like, at 4-steps of 53-EDO, the justly tuned AcA1 and Grm2 are basically identical as humans can discern them: 

    t(AcA1) = 135/128 ~= 92 cents
    t(Grm2) = 256/243 ~= 90 cents
    2^(4/53) ~= 91 cents

So the system is at least finding 5-limit frequency ratios that are indistinguishable and making them identical. But also, a lot of the good simple intervals with good simple ratios are slipping through the cracks if we only use the Turkish steps, [0, 4, 5, 8, 9, 13, ...]. 53-EDO has lots of divisions and some are of course going to be close to simple 5-limit intervals, but the supposedly-Turkish subset on steps [0, 4, 5, ...] doesn't look nearly as good as the full set.

Let's combine the rank-2 and rank-3 analyses!:

0: P1

4: Pythagorean m2, AcA1, Grm2
5: Pythagorean A1, Just m2
8: Pythagorean d3, Just M2
9: Pythagorean M2, AcM2
13: Pythagorean m3, Grm3
14: Pythagorean A2, Just m3
17: Pythagorean d4, Just M3
18: Pythagorean M3, AcM3, Grd4
22: P4
23: Pythagorean A3, Ac4
26: Pythagorean d5, AcA4
27: Pythagorean A4, Grd5
30: Pythagorean d6, Gr5
31: P5
35: Pythagorean m6, AcA5, Grm6
36: Pythagorean A5, Just m6
39: Pythagorean d7, Just M6
40: Pythagorean M6, AcM6
44: Pythagorean m7, Grm7
45: Pythagorean A6, Just m7
48: Pythagorean d8, Just M7
49: Pythagorean M7, AcM7, Grd8
53: P8

So, like, the popular analysis of Turkish microtones by Arel-Ezgi-Uzdilek uses the rank-2 intervals, and notes that there are some intermediate tones between the major Nths and minor Nths. These happen to closely approximate the 5-limit / justly tuned versions of those intervals. And that's it. Turkish music is Pythagorean with Just microtones. One problem with this combined analysis is that e.g. 2^(5/31) is closer to the Pythagorean A1 than to the Just m2, so in a sense the simple Pythagorean analysis has less error. But it's still pretty close. And a merit of the combines analysis as that musicians use 2^(5/31) as if it were some kind of a second interval, so it's nice that we can explain it's second-ordinal function.

...

I read a cool dissertation of Turkish maqamat by Ozan Yarman. He advocates a few analytic systems based around (53 * 3 =) 159-EDO.  He didn't say this, but I'm here to tell you that we can make a 159-EDO by tempering out the the interval dddddddddddddddddddddd5, which has coordinates (-15, 4) in the (A1, d2) basis.

Lol, oh shit, tempering that makes 153 EDO. The actual coordinates of the interavl to temper out are (-21, 1). Which is a dddddddddddddddddddddd2, I think. Because (0, 1) is d2, and then to get -21 on the first coordinated, you just add on 21 more {d}s. Yeah.

Yarman doesn't restrict himself to 24 intervals that are once-modified or natural: he's got names for 79 different intervals within 159-EDO. If you've ever been frustrated that people talking about maqamat are super vague about the true frequency ratios and how they vary by culture and region, look into Ozan Yarman.

I read the paper and I thought it was good enough to share, but I haven't really digested it much. I think I'm going to go through Turkish maqamat first and try to notate them here. I don't know how much they differ in writing from Arabic maqamat that I wrote about in the post on quartertones. Apparently there are subtle differences in performance even between the Arabic versus Turkish maqamat that are written identically. But I'm going to start with the writing and then revisit what Ozan Yarman has to say about the performance after.

...

Okay. I have lots of notes from lots of Turkish PDFs and websites and blurry jpegs. Some of the following might be right. Some of these might be real ascending scales with the correct names:

[K S T S A12 S T]: Karcığar
[K S T T B T T]: Acem or Uşşak or Beyâti
[K S T T K S T]: Nevâ or Hüseynî
[S A12 S T B T T]: Hümâyun
[S A12 S T K S T]: Hicaz or Uzzâl. Might also be called Hicaz Ailesi? 
[S A12 S T S A12 S]: I think this one ascending it's called Zirgüleli Hicaz and descending it's called Hicazkâr. It happens to be a palindrome.
[S T K S T T K]: Irak
[S T K T S A13 B]: Segâh
[T B T T B T T]: Bûselik or Nihavend
[T K S T T K S]: Râst
[T K S T S A12 S] : Basit Sûzinâk
[T S A12 S T K S]: Nikriz
[T T B T T T B]: Çârgâh

Those letters in the brackets indicate numbers of steps in 53-EDO, according to the following mapping:

"F": 1,
"E": 3,
"B": 4,
"S": 5,
"K": 8,
"T": 9,
"A12": 12,
"A13": 13,
.

Usually the "A" sized scale step is 12 steps of 53-EDO but sometimes it's 13, so I've indicated "A12" and "A13" throughout the maqamat. My sources mostly didn't notate this, and I just picked the version that made the octave complete and the whole set of scale intervals normal (i.e. natural or once modified). Also, F and E don't really show up. If pretended that E showed up sometimes, then we can call it BASKET notation. I'll just stick with BSKTA in ascending order of size.

By accumulating a running sum of EDO steps as you go through the list of BSKTA letters, you can get EDO steps for every maqam scale degree, which can then be converted to rank-2 intervals through the mapping:

    edo_step_to_interval = {
0: "P1",
4: "m2",
5: "A1",
8: "d3",
9: "M2",
13: "m3",
14: "A2",
17: "d4",
18: "M3",
22: "P4",
23: "A3",
26: "d5",
27: "A4",
30: "d6",
31: "P5",
35: "m6",
36: "A5",
39: "d7",
40: "M6",
44: "m7",
45: "A6",
48: "d8",
49: "M7",
53: "P8",

This maqam are also well formed ascending scaled, but I've only see n it referenced once:

[T S A12 S S A12 S]: Yegah'da Nev'eser

I think "Yegah'da" just refers to the first note of the maqam as it was written on the staff. If I only present the intervals, its probably just called Nev'eser.

There's a Saba maqam, but it's a little more than an octave and I'm confused where the tonic is. It's sometimes written like this: [K S S A B T S A S]. If we lop off the first letters under the interpretation that they are an ornament below the tonic, then we get [S A13 B T S A12 S], which adds up to 53 steps and produces the normal intervals of [P1, A1, M3, P4, P5, A5, d8, P8]. That might sound kind of arbitrary, especially with how I snuck in both an A13 and an A12, but I tried a ton of things and that's the only one that works. It at least has an interpretation in terms of tetrachords? I think a Turkish music theorist would recognize [S A13 B] and [S A12 S] as two versions of the Hicaz tetrachord / dörtlüsü.

I think Küçek is also more than an octave. Haven't figured it out yet. I also don't understand the Ferahnâk maqam and I'm done trying. I've seen it presented as two runs of notes, each less than an octave, and the notes aren't sorted, and there are differing accidentals if a note shows up more than once.

Some maqamat have a different form when descending:

[-T -B -T -T -S -K -T]: Râst descending
[-T -S -A12 -S -T -B -T]: Bûselik descending

I was silly and my program analyzed those intervals without the negative signs, i.e. as scales going up.  They do work as scales going up, but I haven't checked that they work going down, i.e. that they hit normal intervals . They probably do.

I think you normally go down Çârgâh the way you come up, but there's still a special name for the descending scale? It's Acemasiran or Acem Aşirân: [-B -T -T -T -B -T -T]. The name kind of looks like "the Assyrian form of the Acem makam", right? I don't know. I don't speak Turkish. Also, I think "Mâhûr" might be a synonym for Çârgâh. And I think Kürdî might also be descending Çârgâh? People love to say that although Çârgâh sounds like the western major scale it's not important to Turkish musicians. But they sure seem to have a lot of names for it. Or I'm analyzing things incorrectly.

If you go down Râst the way you came up, it's called Gerdâniye, [-S -K -T -T -S -K -T]. And I think Rast and Gerdaniye are both names for specific pitches in Turkish music, and Gerdaniye is an octave above Rast, so it's kind of a fitting name.

I've seen "Eviç" notated as [S T K S T T K], which is Irak, and as [S T T K S T K]. I've also read the claim that Eviç is the descending form of Irak. I wonder if Irak might refer to Iraq. Stranger things have happened. 

Isfahân is a name for descending Beyâti. Which I think is also Acem and Uşşak. Maybe those three aren't all the same and I've made some mistakes. But I still pretty sure that Isfahân is [-T -T -B -T -T -S -K] instead of [K S T T B T T].[0, 5, 14, 19, 31, 36, 49, 53]

I've seen this as a variation on Râst: [T K S T T B T]. It might be called Acem'li or Pençgâh? I've also seen the reverse of that called Acem'li rast descending, [T B T T S K T].  And I've also seen [T S K T S K T] called Acem'li descending. I'm pretty lost.

There are supposedly hundreds of maqamat.  I wish someone would make a machine-readable list. I'm doing my best. Hereafter are the ones I want to check my notes and the internet for to see why I don't have them listed above, because I thought I'd taken notes on them: [Büzürk, Hisar, Müsteâr, Nişâbûr, Rehâvi, Zirefkend]. Also, is there a plain Zirgüle besides Zirgüleli Hicaz?

Sometimes you'll see the claim that maqamat are made of a tetrachord (four notes spanning P4, not actually played together harmonically as a chord), and a pentachord (five tones spanning P5, usually just made from a P4 tetrachord plus a major second to reach P5). From what I've seen, only like half of the maqamat are analyzable this way. But maybe I have an impoverished set of tetrachords in my analytical library.

Once I've re-researched the key missing maqamat, I'll compare what's here in 53-EDO to Ozan Yarman's work in 159-EDO.

The Hüzzâm maqam is this: [S T S A12 S A13 B]. And it's frustrating to me. Here it is as steps of 53-EDO from P1 to P8: [0, 5, 14, 19, 31, 36, 49, 53]. The 19 is the frustrating part. That step doesn't correspond to one of our normal intervals. It's not in [0, 4, 5, 8, 9, 13, 14, 17, 18, 22, 23, 26, 27, 30, 31, 35, 36, 39, 40, 44, 45, 48, 49]. No, 19 steps of 53-edo as a rank-2 interval is AA2. Twice modified. I wish this was a typo? But everyone agrees Huzzam is [S T S A S A B], and there isn't an assignment of 12 and 13 steps to the As that makes it any better. I don't get it.

Ali C. Gedik posted some turkish maqamat in terms of 53-EDO steps in their PhD thesis, "Automatic Transcription Of Traditional Turkish Art Music Recordings".

These ones match what I have above:
    rast = [0, 9, 17, 22, 31, 40, 48, 53]
    ussak = [0, 8, 13, 22, 31, 35, 44, 53]
    nihavend = [0, 9, 13, 22, 31, 35, 44, 53]
    huzzam = [0, 5, 14, 19, 31, 36, 49, 53]
    huseyni = [0, 8, 13, 22, 31, 39, 44, 53]

And these ones do not:
    hicaz = [0, 5, 17, 22, 31, 35, 39, 44, 53]
    segah = [0, 5, 14, 22, 31, 36, 45, 49, 53]
    kurdili_hicazkar = [0, 4, 13, 22, 31, 35, 44, 53]
    saba = [0, 8, 13, 18, 31, 35, 44, 49]
.
Hicaz and Segah have one more note here than I expected. Also, Saba doesn't have the octave. I hadn't heard of kurdili hicazkar. The Turkish wikipedia article on it makes it sound like a real banger: popular, uplifting, some kind of super group mecha makam made of three smaller ones but still having the regular number of scale degrees.

I think it's

    Kürdi’li Hicazkâr Makamı: [-T -T -B -T -T -T -B]

And sometimes there's an extra [-T -B -T] on top at the start as an ornament.

Here's a classification of maqamat from the Turkish wikipedia:

    Simple: Çargah, Bûselik, Rast, Uşşak, Hicaz, Uzzal, Hümâyun, Zirgüleli Hicaz, Neva, Hüseynî, Karacığar, Suzinak, Kürdî

    Migrated (sed): Acemaşiran, Mahur, Sultaniyegah, Nihavend, Kürdilihicazkar, Zirgüleli Suzinak, Şedaraban, Evcara, Hicazkar, Suz-i dil, Ruhnevaz, Ferahnüma, Aşkefza, Heftgah

Let's check that we have all of those.

...

Or! Or I could go back to the title of the post. Let's rationalize 53-EDO. Based on the math in the next post ("Tempered Frequency Ratios? Why, I never!"), here are the justly tuned frequency ratios for rank-5 (11-limit) intervals and their number of steps in 53-edo:

0 - 1/1
1 - 50/49, 55/54, 64/63, 81/80, 100/99
2 - 33/32, 36/35, 45/44, 49/48, 56/55, 77/75
3 - 22/21, 25/24, 28/27, 80/77
4 - 21/20, 81/77
5 - 15/14, 16/15, 35/33, 77/72
6 - 27/25, 88/81
7 - 11/10, 12/11, 35/32, 49/45, 54/49
8 - 10/9, 55/49
9 - 9/8, 28/25, 49/44
10 - 8/7, 25/22, 55/48, 112/99
11 - 63/55, 81/70
12 - 7/6, 33/28, 64/55, 75/64, 88/75, 90/77
13 - 25/21, 32/27
14 - 6/5, 77/64, 105/88
15 - 11/9, 40/33, 60/49, 98/81, 121/98
16 - 27/22, 49/40, 99/80, 100/81
17 - 5/4, 44/35, 56/45, 96/77, 121/96
18 - 63/50, 80/63, 81/64, 125/98, 125/99
19 - 9/7, 14/11, 32/25, 77/60
20 - 35/27, 55/42, 64/49, 100/77, 125/96, 128/99
21 - 21/16, 33/25, 72/55, 98/75
22 - 4/3, 66/49, 75/56, 121/90
23 - 27/20, 110/81
24 - 11/8, 15/11, 48/35, 49/36, 135/98
25 - 25/18, 88/63, 112/81
26 - 7/5, 45/32, 99/70, 108/77
27 - 10/7, 64/45, 77/54, 121/84, 125/88, 140/99
28 - 36/25, 63/44, 81/56
29 - 16/11, 22/15, 35/24, 72/49
30 - 40/27, 81/55, 121/81, 125/84
31 - 3/2, 49/33, 112/75, 121/80
32 - 32/21, 50/33, 55/36, 75/49
33 - 49/32, 54/35, 77/50, 84/55, 99/64, 125/81, 135/88
34 - 11/7, 14/9, 25/16, 120/77
35 - 63/40, 100/63, 128/81
36 - 8/5, 35/22, 45/28, 77/48, 121/75
37 - 44/27, 80/49, 81/50, 125/77, 160/99
38 - 18/11, 33/20, 49/30, 81/49, 105/64
39 - 5/3, 121/72, 128/77, 165/98
40 - 27/16, 42/25, 147/88
41 - 12/7, 55/32, 56/33, 75/44, 77/45, 121/70, 128/75
42 - 110/63, 125/72, 140/81
43 - 7/4, 44/25, 96/55, 99/56, 135/77
44 - 16/9, 25/14, 88/49, 175/99
45 - 9/5, 98/55
46 - 11/6, 20/11, 49/27, 64/35, 90/49, 175/96
47 - 50/27, 81/44, 147/80
48 - 15/8, 28/15, 66/35, 121/64, 144/77
49 - 40/21, 121/63, 125/66, 154/81
50 - 21/11, 27/14, 48/25, 77/40
51 - 35/18, 55/28, 64/33, 88/45, 96/49, 125/64, 150/77
52 - 49/25, 63/32, 99/50, 108/55, 125/63, 160/81
53 - 2/1
.

...