Maqam Sikah Baladi

Maqam Sikah Baladi is an Arab maqam (rather than Turkish or Persian) with a lot of microtones. We're going to try figuring out some things about it.

MaqamWorld decsribes an associated scale fragment, jins Sikah Baladi, as:

    [Ed, Ft, G, Ad, Bd, C] 

Those are the base pitch classes that would indicate 24-EDO steps, but they have some extra accidentals that I haven't transcribed.

The Ed is flatter than 24-EDO, the Ft is sharper, the Ad is flatter, the Bd is sharper, and the C is flatter.

If we ignore those extra accidentals, I would notate this in 24-EDO steps as:

    [4, 3, 3, 4, 3]\24

If we include the accidentals, we'll have to commit to some kind of step size for the flattening and sharpening. Let's arbitrarily call it 0.25 steps. If it was more than 0.5 steps, then we'd be closer to the next step and the pitches would have been notated differently. With that intonation, we get this for our steps:

    [4.5, 2.75, 2.75, 4.5, 2.5]\24.

MaqamWorld says that the tonic/finalis is on G, and that there is no ghammaz.

Oud player Joseph Tawadros describes maqam sikah baladi as 

    [C, Dd, Ed, Ft, G, Ad, Bd, C]

in a Facebook video. The jins sikah baladi from MaqamWorld is a subset of this, so that's a nice bit of agreement. In 24-EDO these pitch classes are:

    [0, 3, 7, 11, 14, 17, 21, 24] // relative

    [3, 4, 4, 3, 3, 4, 3] // absolute

with the initial [3, 4] being the relative intervals not accounted for in the jins. And those are the intervals of the normal jins Sikah, so everything looks good so far.

Now let's compare to maqam Sikah Baladi from MaqamWorld. They have it written descending, and I'm sure it's played in a descening way, but I like to write my scales ascending and will do so now.

The maqam starts low with part of jins Sikah baladi (starting in te middle at the tonic on G) and moving up tot that flat C like beofre:

    [G, Ad, Bd, C]

Then there's a C# as part of jins "pseudo-hijazkar / suspended 5th"

The pitch classes of this are

    [C#, D, Ed, Ft, G]

From the pitch classes, we'd expect this to have an 24-EDO intonation like

    [2, 3, 4, 3]

Agai    n this has non-24-EDO accidentals, indicating a flat Ed and a sharp Ft. If we use the 0.25 step intonation again, then we get

    [2, 2.75, 4.5, 2.75]

Let's compare this jins "pseudo-hijazkar / suspended 5th" to regular jins hijazkar.

MaqamWorld presents jins Hijazkar in terms of major tones as

    [3/2, 1/2, 1/2, 3/2] tones

i.e.

    [6, 2, 2, 6]\24

with pitches

    [Ab, B, C, Db, E, F] 

and notes that the C in the middle is the tonic and there is no ghammaz. That looks absolutely nothing like pseudo-hijazkar.  It doesn't even have neutral tones. I think we just have to ignore that jins label.

Let's continue on with the maqam from MaqamWorld. Overlapping with jins pseudo-hijazkar we have another jins Sikah Baladi, this time the standard one that stretches from a flat Ed up to a flat C. I will note that there are numbers 1 through 7 starting under the D and moving up to the high C, as through D is the tonic of the whole maqam and it doesn't reach the octave, and the C# in the middle is just a leading tone up to the D.

I'm tempted to ignore the low jins Sikah Baladi since 

    1) it doesn't give as any new information (being repeated above),   

    2) it messes up the intervallic structure when we have both C natural and C#, and 

    3) The tetrachord doesn't even present in full, only including from G up to C instead of from Ed up to C.

If we skip all the notes from the lowest jins, then our maqam Sikah Baladi loos like this:

    [(C#), D, Ed, Ft, G, Ad, Bd, C]

Using the same intonation for the non-24-EDO accidentals, this would look like 

[(2), 2.75, 4.5, 2.75, 2.75, 4.5, 2.5]\24.

in relative steps, which is an adjusted version of this:

[(2), 3] + [4, 3, 3, 4, 3] \24

Across the whole maqam, including the lower jins Sikah Baladi, the notes which are highlighted as special targets for tonicization and/or ghammazization are [G, Bd, D, G, Bd]. So I don't know why people ever present the scale spanning from C to C.

The maqam Sikah Baladi of Tawadros reaches the octave and has Dd instead of D, but is otherwise quite close.

If we altered the MaqamWorld version of the maqam so that we had a tonic of a 0.25-step-flattened C natural (instead of a leading tone of C#), then the maqam would be:

    [4.25, 2.75, 4.5, 2.75, 2.75, 4.5, 2.5]\24.

which does reach the octave. This still differs from Tawadros's maqam in having D natural instead of Dd, but there's more agreement than disagreement.

Navid from OudForGutarists posted a piece in Sikah Baladi with key signature [Ad, Ed, Bd, Ft], which is consistent with these pitches:

    [C, D, Ed, Ft, G, Ad, Bd, C]

He starts his phrases on Ed, works up to a few notes to G, and ten works down to a low G where he ends his phrases, so it's more like

    [G, Ad, Bd, C, D, Ed, Ft, G]

Anyway, with two sources agreeing on D instead of Dd, I'm going to go with that.

One more data source? There are some video on youtube from user @FantasticoTube that maqamat and look like they're from a website that has gone defunct. Honestly, they look like an early version of MaqamWorld, but perhaps they're unrelated. The FantasticoTube video presents Sikah Baladi twice. The first form looks like this:

    [G, Ab, B, C, D, Eb, F#, G]

    [1/2, 3/2, 1/2, 1, 1/2, 3/2, 1/2] tones

with arrow accidentals indicating that Ab is played sharper, B is played flatter, Eb is played sharper, and F# is played flatter. Tetrachord annotations describe it as 

    [Hijaz on G + major second + Hijaz on D]

The second form looks like this:

    [G, Ad, Bd, C, D, Ed, Ft, G]

   [3/4, 1, 3/4, 1, 3/4, 1, 3/4] tones

It is annotated with jins as 

    [Sikah on G] + 3/4 + 1 + 3/4 + [Sikah on D]

This version with quarter tones and jins Sikah does indeed move the maqam in the direction indicated by the arrow accidentals from the first form.

So the idea seems to be to alter jins hijaz

    [2, 6, 2]\24

by quarter tones to give 

    [3, 4, 3]\24

which has its first wo intervals the same as a 24-EDO jins Sikah, although this guy here is an actual tetrachord whereas the traditional jins Sikah is just a trichord.

Anyway, the second form form of maqam Sikah Baladi from FantasticoTube has the same pitch classes and arrangement as my summary of Navid's maqam Sikah Baladi ending on G:

    [G, Ad, Bd, C, D, Ed, Ft, G]

So now we've got tons of agreement. Here it is in integer steps of 24 EDO:

    [3, 4, 3, 4, 3, 4, 3]\24

Using the sub-quarter tone intonation inspired by notes from MaqamWorld (still with 0.25 steps unless you've got a better idea), this is our more precise intonation given this arrangement of pitch classes:

    [G (2.75) Ad (4.5) Bd (2.5) C (4.25) D (2.75) Ed (4.5) Ft (2.75) G]

The 24-EDO version does indeed have a repeated structure with tetrachords that are like jins Hijaz adjusted toward jins Sikah:

    [3, 4, 3] + T + [3, 4, 3]

Although the repeated structure is less exact in my intonation with the made up 0.25 tone adjustments.

    [2.75, 4.5, 2.5] + [4.25] + [2.75, 4.5, 2.75]

I think this is a slight problem. If we adjust the intonation so that the lower group of three relative intervals sum a tempered P4 at 10 steps of 24-EDO, then we get C natural instead of the slightly flat C that MaqamWorld specifies. (The upper group of three relative intervals is already a true tetrachord spanning 10 steps).

I think we need more data to make more conclusions.

If we want to make a just tuning based on the 0.25 step intonation, here's one option:

[P1, Prm2, ReM3, HbAc4, P5, Prm6, ReM7, P8] # [1/1, 13/12, 16/13, 45/34, 3/2, 13/8, 24/13, 2/1] _ [0, 139, 359, 485, 702, 841, 1061, 1200] cents

[Prm2, ReReAcA2, HbPrAcm2, ExM2, Prm2, ReReAcA2, Prm2] # [13/12, 192/169, 585/544, 17/15, 13/12, 192/169, 13/12] _ [139, 221, 126, 217, 139, 221, 139] cents

It looks pretty good in absolute intervals and a little crazy in relative intervals. If we assume adjustments more like 12 cents, then this is a decent tuning:

[P1, AsGrm2, GrM3, HbAc4, P5, AsGrm6, GrM7, P8] # [1/1, 88/81, 100/81, 45/34, 3/2, 44/27, 50/27, 2/1] _ [0, 143, 365, 485, 702, 845, 1067, 1200]

[AsGrm2, DeAcA2, HbAcAcm2, ExM2, AsGrm2, DeAcA2, Acm2] # [88/81, 25/22, 729/680, 17/15, 88/81, 25/22, 27/25] _ [143, 221, 120, 217, 143, 221, 133]

Although I kind of doubt that the maker of MaqamWorld would even write a note about a difference of intonation as small as 12 cents.

We need more data to draw more conclusions.

...

OffTonic Theory! It's a website where a guy describes the use of Arabic maqamat in Syrian Jewish liturgical music. And the author of the site uses 53-EDO instead of 24 EDO. He analyzes the maqam as having a hijaz-like tetrachord at the base of either [6, 10, 6] or [7, 8, 7] relative steps. A frequency ratio between 8 and 10 steps of 53-EDO isn't all that precise - it's a 45 cent difference, but we can use this. We are not restricted to using integers. What's the intonation of Sikah Baladi? About [6.5, 9, 6.5], i.e. [147, 204, 147] cents relative or [0, 147, 351, 498] cents absolute. Those are actually surprisingly close to 24-EDO values of [0, 150, 350, 500] cents absolute or [3, 4, 3]\24 steps relative. So a 24-EDO intonation isn't too bad among Syrian Jews at least. A nice nearby just intonation for this is

    [1/1, 12/11, 27/22, 4/3] absolute

    [12/11, 9/8, 88/81] relative

A this one's a less simple but it's symmetric:

    [1/1, 12/11, 11/9, 4/3] absolute

    [12/11, 121/108, 12/11] relative

So there's another option for intonation, though I can't help but feel that the intonation I made up is better.

...

Ah! Got it.

Sami Abu Shumays gives a .scl file for a pitch set that includes Jiharkah and its modulations and says that jins Sikah Baladi is an option: https://tuning.ableton.com/arabic-maqam/jiharkah/. The tricky bit is that this Sikah Baladi has its tonic on C instead of G, so we'll have to transpose.

Here's our maqam Sikah Baladi from MaqamWorld:

    [G, low Ad, high Bd, low C, D, low Ed, high Ft, G]

Move that up P4 or down P5 and we get:

... nothing. Those pitches aren't in his pitch set. We'd be a lot closer if we hadn't transposed.

Rooting the maqam on G and using the closest available pitches for modulation, we get [G, Ab, Bd, C, D, E-, F#, G] -> [0, 123, 359, 498, 707, 876, 1088, 1200]. This looks plausible as an intonation (even if the pitch names aren't all correct) everywhere except for the E-, which should instead be a low E half flat instead of a low E natural. A low E half flat should be like 310 to 340 cents over C natural / perde Rast, so like 808 to 838 cents over G.

...

Oh, oops, he wasn't saying that maqam Sikah Baladi was a modulation option, just jins Sikah Baladi. So I shouldn't be looking for the full set of tones in the maqam. His intonation for just the jins  is: 

...

...

I asked the xenharmonic discord about the intonation of Sikah Baladi and Margo Schulter had an interesting take.

"The traditional interpretation might be like the versions of Systematist Buzurg from around 1300"

[1/1, 14/13, 16/13, 4/3] _ [0, 128, 359, 498] cents

[14/13, 8/7, 13/12] _ [128, 231, 139] cents

or

[1/1, 13/12, 26/21, 4/3] _ [0, 139, 370, 498] cents

[13/12, 8/7, 14/13] _ [139, 231, 128] cents

Buzurg (also called Buzurk, Buzrak, and Bozorg) is an difficult and interesting thing to pin down in the history of middle eastern music, but Margo Schulter is cool, so let's try.

Buzurg is, in one sense, a pitch or a perde or a place on the neck of a lute like a tanbur or oud. It's specifically an octave above perde Sikah. Thus if perde Rast is called C, then Sikah and Buzurg are both Ed, in different octaves.

Buzurg was also a mode in medieval Persian and Arabic music. My understanding is that Safi al-Din described the mode as 

    [P1, GrGrGrd3, GrGrd4, P4, GrGrGrd6, P5, AcM6, GrGrd8, P8] # [1/1, 65536/59049, 8192/6561, 4/3, 262144/177147, 3/2, 27/16, 4096/2187, 2/1] _ [0, 180, 384, 498, 678, 702, 906, 1086, 1200] cents

which is ugly due to being Pythagorean, but we give it a schismastic reinterpretation as:

    [P1, M2, M3, P4, Gr5, P5, AcM6, M7, P8] # [1/1, 10/9, 5/4, 4/3, 40/27, 3/2, 27/16, 15/8, 2/1]

    [M2, AcM2, m2, Ac1, AcM2, M2, m2] # [10/9, 9/8, 16/15, 10/9, 81/80, 9/8, 10/9, 16/15]

This is much nicer looking, it's spelled correctly intervallically (except for having two 5th intervals), and it's aurally indistinguishable from the previous intonation.

It's also basically just a major scale (with a weirdly doubled up 5th scale degree).

Encyclopedia Iranica says that the later Qoṭb-al-Dīn Šīrāzī in his "Dorrat al-tāj" provided corrections to Safi al-Din's work and gave a jins for Bozorg as a pentachord:

Absolute: [G, Ad, B, C, C#(+), D] _ [0, 150, 417, 498, 626, 702]

Relative: [150, 267, 81, 128, 76]

which can be extended with these tones:

Absoltue: [E, Gb, G] [204, 386, 498]

Relative: [204, 182, 112] cents

To give this full scale:

Full scale:

[G, Ad, B, C, C#(+), D, E, F#, G]

[0, 150, 417, 498, 626, 702, 906, 1088, 1200]

This has a fairly obvious detempering:

Absolute: [1/1, 12/11, 14/11, 4/3, 56/39, 3/2, 27/16, 15/8, 2/1]

Relative: [12/11, 7/6, 22/21, 14/13, 117/112, 9/8, 10/9, 16/15]

Although it's not spelled correctly in my interval naming system (with the 7/6 being a 3rd interval and the 22/21 being a 1st, though we'd hope for them to both be kinds of 2nds in order to get a scale that progressed alphabetically). That's fine. Not every medieval middle eastern music theorist will use my interval naming system.

The reference for that intonation in Encyclopedia Iranica is from Owen Wright. This definitely looks closer to a distinct middle eastern mode than Safi al-Din's garbage major scale, and it's closer to Margo Schulter's jins. I think both Safi al-Din and Qutb al-Din al-Shirazi have been described as being Systematists about middle eastern modes, not that I really know what that means (although I think they're the ones who introduced a [T B J]-like notation for tetrachords, and they might have championed 1/3 tones in contrast to quarter tones, and are they're probably involved any time you see an intonation involving 8/7 or 7/6), and they both lived around 1300 AD, so we are looking at "versions of Systematist Buzurg from around 1300", if you were curious.

Encyclopedia Iranica also lists some modern things called Bozorg across various middle eastern musical traditions. It's a shashmaqam in Tajikistan and Uzbekistan, "characterized by a pentatonic structure: (C) D E (F#) G A B (C) D." I don't know exactly what they mean by that, but it sure doesn't look anything like the microtonal scale of Qutb al-Din al-Shirazi. In modern Persian music, Bozorg is a melodic motif played in Dastgah Shur that goes through these notes [C, Dp, Eb, F, G, (Ap | Ab), Bb] but in a broadly descending order. This at least has a microtone or two. And it's also some other things. I don't know how these are connected and neither does Encyclopedia Iranica. It doesn't seem to be that they all emphasis a note that's an octave and a neutral third above Rast.

In modern Turkish music, Büzürk/Büzürg is a compound maqam described in ascending order as a "Buselik pentachord on A, a Huseyni pentachord on E, and a Çargâh pentachord on G." I'm sorry to say this is just Arel-Ezgi-Uzdilek bullshit and probably it can't tell us anything about the history of Buzurk, but let's try.

Genus Buselik is jins Nahawand, i.e. [9/8, 256/243, 9/8], and we can add on one more 9/8 for a pentachord. Huseyni is also called Ussak in Turkish music theory, but in Arabic it's called Bayati and has an intonation like [13/12, 128/117, 9/8] or [88/81, 12/11, 9/8]. Add on another 9/8 to get a pentachord. Finally Çargâh is the must infuriating name in all of the AEU corpus, because it's a Persian name for a tetrachord like Hijaz but they use it for Ajam, because they wanted a historic name for the major scale and to pretend like the western major scale was ever of central importance to their music, and their solution for that was .... to ignore history and change their own historic names. They already has Acem as a perde that was cognate with 'ajam. I get so mad when I see this, Hüseyin Sadeddin Arel. I curse you in your grave.

Anyway, jins 'Ajam is [9/8, 9/8, 256/243], and we can add on 9/8 for a pentachord. This doesn't look like any other Buzurg to my eye. I don't think it really even deserves to be called a "compound" maqam. It's just a seyr.

So, Buzurg. Is it related to Sikah Baladi? Yeah. Margo Schulter's 

    [14/13, 8/7, 13/12] and [13/12, 8/7, 14/13]

are both tuned to [6, 10, 6] steps of 53-EDO, like one of OffTonic Theory's intonations. It definitely looks like a modified jins Hijaz, and modified in the right direction.

Unlike my detempering of Bozorg from Qoṭb-al-Dīn Šīrāzī, Margo Schulter's tetrachord is spelled by 2nd intervals, which I really like. I think I slightly prefer her intonation with a leading 13/12, just based on it's similarity to my theoretical intonations for Sikah Baladi. If we make a maqam out of it, repeating the tetrachord with a AcM2 disjunction, we get:

    Absolute: [P1, Prm2, PrSpGrm3, P4, P5, Prm6, PrSpGrm7, P8] # [1/1, 13/12, 26/21, 4/3, 3/2, 13/8, 13/7, 2/1] _ [0, 139, 370, 498, 702, 841, 1072, 1200] cents

    Relative: [Prm2, SpM2, ReSbAcM2, AcM2, Prm2, SpM2, ReSbAcM2] # [13/12, 8/7, 14/13, 9/8, 13/12, 8/7, 14/13] _ [139, 231, 128, 204, 139, 231, 128] cents

...