Mike's Oud Forums - Powered by XMB

Oud Fingerboard 'Fret' Inlays?

jdowning - 7-5-2010 at 12:28 PM

The fingerboard of my old 'Egyptian' oud has inlays that seem to represent 'frets' or - more correctly - finger positions as far as a modern oud (19th C onwards) is concerned. On the other hand these inlays may be purely decorative although , perhaps, a dim and distant memory of times when the oud was fretted?

In the topic "Nahat in Brazil" ('Oud Forum') two images of ouds with 'fret' inlays have been posted. The first of an oud by Abdo George Nahat (1959) and the second by a currently unknown luthier Emil Khoury of Beirut, Lebanon. The latter instrument apparently was owned by the late (and lamented) oud virtuoso Farid Al-Atrash.
The image of the fingerboard of the Emil Khoury Oud - although only a low resolution 'thumbnail' - was sufficient to allow an analysis. Assuming the distance from the front edge of the nut to the neck joint is 1/3 string length (as it should be) the attached image compares relative 'fret' or finger position calculated according to 'Western' 12 Tone Equal Temperament and 'Pythagorean'. It can be seen - within reasonable tolerance errors - that the inlays are intended to represent fret or finger positions and so are not purely decorative.
I shall next analyse my old 'Egyptian' oud finger board to see if there is any supportive correlation.

If the fingerboard inlays do represent finger positions, is this an indication of a traditional regional design (Lebanese or neighbouring regions)?
I ask this question in the light of the fingerboard inlay by another (unknown) Lebanese luthier (image attached).
Curiouser and curiouser - more to follow!

Emil Khoury Fingerboard (421 x 847).jpg - 85kB

fernandraynaud - 7-5-2010 at 02:10 PM

I too have an Egyptian with markers. They seem intended as chromatic position lines, but sloppily measured.

The free WFret program will nicely calculate accurate positions for any scale length and number of "frets". It will print a nice template. I have used these, with a felt-pen, to dot accurate reference markers on the strings once they have stretched, for fretless bass and on one oud.

ALAMI - 7-6-2010 at 10:11 AM

The Sumbat Bedrosian I have has also these frets-like inlays

http://www.mikeouds.com/messageboard/viewthread.php?tid=7968#pid503...

They may be helpful for beginners or just decorative, most of the modern oud makers I've met, excepted Nazih Ghadban, were not even aware that ouds were fretted back in time.

jdowning - 7-6-2010 at 12:42 PM

Thanks Fernandraynaud and ALAMI.
I have not tried WFret but suspect that it will only calculate fret positions according to Western 12 Tone Equal Temperament (12 TET). I am interested also - but more so - in the historical fret positions for European instruments as well as those given by the Arab and Persian theorists of earlier times for the oud.
I am currently using a useful spread sheet for the European historical fret placements available as a free download from the Lute Society of America at
http://www.cs.dartmouth.edu/~lsa/
Check the 'Downloads' button.
With this software it is only necessary to input string length to obtain fret positions for 16 historical placements - including modern Western 12 TET and Pythagorean (used by the early Arab and Persian theorists).
For the historical oud fret placements it is necessary for me to then go through some additional calculations. To enable relative comparisons to be made between 'East and West' I must then convert to conventional 'cents' (1 cent = 0.01 semitone TET).

The 'fret' inlay examples posted (decorative or 'true'?) are obvious, however, more subtle examples (yet to be investigated) might be hidden in the 'Arabesque' fingerboard inlays more commonly found where a pattern is repeated (or regularly terminated) possibly at the 'fret' or finger position for the particular regional musical scale familiar to the luthier? These patterns may have been handed down through successive generations of luthiers but without present day oud makers being conscious of the historical significance (if there is any) of the inlays.

Analogous to the 'mode' (maqamat) calligraphy carefully and artistically hidden in the rosette patterns - might also be the required finger positions on the fingerboard. All very familar to a professional oud player but also a useful reminder to the beginner oud player as well no doubt?

Of course, if the inlays (or pattern repeats) do 'precisely' signify finger positions, the tolerance for exact location would be somewhat greater than for a tied fret position.
More to follow.

fernandraynaud - 7-6-2010 at 10:19 PM

I am familiar with, and personally use various "temperaments", in tuning keyboard instruments, where each string is independent. And look at this:

http://www.youtube.com/watch?v=t7Cq3pbcMkI&feature=related

I don't fully understand how an instrument with normal frets can really make use of alternate temperaments, as the frets are in a line and affect the whole "row". If we adjust the frets for a pythagorean or meantone temperament for the Neva G string, I don't think we get anything useful for the D string. Can you explain?

There are people making guitars with staggered frets, to address this.

http://www.youtube.com/watch?v=E0pUt_Wvht0&feature=related

http://www.youtube.com/watch?v=YEOY3SqxMc0&feature=related

And we have this:

http://www.erhanbirol.info/tolgahan/EN/microtonal_en.html

But it seems that unless you do something like that, the best compromise is 12ET.

In markers I find the 12ET useful as a base reference, e.g. to play in ET intonation or in a given Maqam and knowing the proper "fudge" say for the third degree of Rast vs. Ajam, and knowing where other instruments will fall. I can't imagine starting from some other "fretting". What am I missing here?

As for elaborate fingerboard inlays of course they are helpful in themselves, as "reminders", as are edge decorations. It would be fascinating if some more explicit meaning were present.

jdowning - 7-7-2010 at 11:22 AM

My interest in this thread is to find out if the fingerboard inlays sometimes found on surviving ouds might be be related to any of the early fretting systems recorded in the early texts as being used on both ouds and lutes
(9th to 17th C).
The early fretting systems were many and varied and included variations of 12 TET placement (i.e. for lutes). For ouds, the early Arab and Persian theorists used fretting systems based on Pythagorean placement.
It should be born in mind that tied gut fretted instruments of earlier times (which I am interested in) are quite different from modern wire strung guitars with immovable metal frets (which I am not interested in - particularly the virtually 'unplayable' so called Just intonation intonation instruments).

I have discussed and compared early oud and lute fretting systems in some detail in my 'Old Oud' project on this forum so am reluctant to go over the same ground again but suffice to say that tied gut frets can be (and were) moved around to suite the mode of the music and the ear of the performer. Gut frets can even be canted at an angle if necessary.
For a lute player performing the polyphonic repertoire of the 16th and 17th C. the starting point for positioning frets might be say - 12 TET (which is my starting point as a lute player) but due to use of varying diameters of gut frets, varying string tensions and string diameters from treble to bass (particularly if the thicker gut or silk basses are used) the frets then require repositioning until everything sounds in tune to the player (which it usually does not - to my ear - if strict 12 TET fret positioning is maintained). The Spanish vihuela composers of the 16th C sometimes instructed a player to reposition certain frets according to the mode of a piece - so that it would sound in tune.
So - yes - fret placement for historical instruments is a compromise (dependent upon a variety of factors) and - no - 12 TET unmodified is less than satisfactory.

jdowning - 7-7-2010 at 12:33 PM

The fingerboard of my old 'Egyptian' oud is another example of 'fret position' inlays. This instrument - currently being restored (as time permits) is interesting. Now battered and worst for wear it was made - according to the roughly made label inside - by Hosan Muhi Al-Din Helmy a 3rd year apprentice attending a (so far unidentified) training school in an unknown location. The oud is assumed to be Egyptian because I bought it in Cairo in the early 1960's - but it could be from anywhere in the Middle East. It is of unknown age (1920's ?).
The current string length is 622 mm. The fingerboard length (front edge of nut to neck joint) measures 202 mm, from neck joint to centre of sound hole measures 212 mm and from sound hole centre to front of bridge measures 207 mm. Front edge of bridge to the bottom of the bowl measures 90 mm. The length of the bowl from neck joint is 511 mm. Due to repairs over time at the neck joint the neck would appear to be a little shorter than original.
As can be seen from the dimensions above, the oud does not meet the traditional proportions of fingerboard length being 3X string length - perhaps due to careless workmanship by the apprentice luthier or perhaps due to a modification of the instrument to increase the string length to 622 mm. At this string length, the fret inlays make no sense.
However, if it is assumed that the original length of the fingerboard was 204.4 mm (I have messed around with several alternatives to arrive at this - not unreasonable assumption) then the original string length with this dimension (according to tradition) should be 3X 204.4 mm = 613.2 mm (say 613 mm). This does not work either because the bridge is too low. However, if we position the bridge according to tradition (half fingerboard length) then things start to come together and the 'fret' inlays start to make some kind of sense.
The attached image compares the inlays to scale against a
12 TET fret positioning and Pythagorean.
It is clear from this that there is a pretty good match with 12 TET particularly if the angling of the inlays (higher on the bass side) is taken into consideration. Therefore, the fret inlays are not randomly placed (and so are unlikely to be purely decorative) and could be intended to represent a slightly modified 12 TET fret placement.\
Is this evidence of a much earlier tradition (now forgotten in the oud making tradition) from a time (15th C or earlier) when oud design had a direct influence lute design in the West? Does this, in turn, mean that my "Egyptian" oud was not made in Egypt but - perhaps in Lebanon? Hard to say without further data.
I shall now check the fret inlay placement of this oud against a lute that has a modified 12 TET fret placement.
More to follow!

Egyptian Oud Fingerboard Fret Positions (474 x 945).jpg - 103kB

jdowning - 7-8-2010 at 10:58 AM

When re-fretting a lute (it has to be done fairly regularly as the gut wears out) I use a wooden gauge stick - marked in 12 TET spacings - to initially position the frets. The frets are then adjusted as required - as the lute is played - in order to bring everything 'in tune'.
Note that some lutenists today work with temperaments other than 12 TET for fret placement to obtain best results.

I have never bothered before to check the deviations of the fine tuned frets from the exact 12 TET spacings so the fret positions have been measured as closely as possible (to about 0.25 mm)with an accurate steel ruler. The string length of the instrument is taken as 602 mm which allows for a slight shortening of each string at the bridge due to the tied 'knot'. Actual string lengths vary from 603 mm at the treble to 601 mm at the bass. This is due not only to the treble and bass strings being more widely separated at the bridge compared to the central third course but also because a lute bridge is made wider and deeper at the bass than at the treble side and is canted towards the nut.
Also the fret diameters reduce in diameter from nut to neck joint in 0.5 mm steps starting at 0.85 mm for fret 1 reducing to 0.50 mm for fret 8
These variations also play a part in the final positioning of the frets.

For information, here are the measured deviations from exact 12 TET - small variations but important.
Fret 1 - 1.2 mm displaced towards bridge
Fret 2 - 1.7 mm '' nut
Fret 3 - 0.3 mm '' bridge
Fret 4 - 0.7 mm '' nut
Fret 5 - 0.5 mm '' nut
Fret 6 - 1.3 mm '' nut
Fret 7 - 0.0 mm '' -
Fret 8 - 0.7 mm '' bridge

These are somewhat similar deviations seen in the measured 'Egyptian' oud fingerboard inlays previously posted (except that they are all displaced slightly towards the nut).
So the conclusion is - having compared the results with 15 other historical fret placements - that the inlays are intended to represent 12 TET spacings.

Fret spacing (used by many guitar makers today) is often calculated according to the so called "Rule of 18" (or more accurately 17.817) which gives the 12 TET fixed fret positions for a given string length. So the first fret position from the nut is found by dividing the vibrating string length by 17.817 which - for my lute works out as 33.8 mm. The string length remaining (i.e measured from the first fret position) is again divided by 17.817 to give the second fret position measured from the first fret as 31.9 mm or, alternatively 65.7 mm from the nut. And so on.
This method, however, is not a modern invention. It's discovery in the year 1591 is attributed to the renowned lutenist Vincenzo Gallilei (the astronomer family) who is said to have invented it after reading the theoretical texts of the early Greeks. Or did he acquire this information from another source - the Arabs or Persians for example - who had translated the same Greek texts some centuries prior to the end of the
16th C?
I have not come across any surviving lutes of the 16th/17th with this type of fingerboard inlay although it would have had a useful, practical application in aiding the preliminary positioning of gut frets for the majority of lutenists - amateur or professional - who, otherwise, would have had no idea how to calculate fret positions mathematically.

As"it takes more than one swallow (or two) to make a Summer" more data is needed for further analysis. Are these two fret boards the only ones in existence that have this type of inlay? More examples are needed so if anyone has an oud with finger board inlays of this kind please post an image on this thread. Thanks.

fernandraynaud - 7-9-2010 at 10:00 PM

Ok, here is my preciouss MOP Egyptian. You can see the deviations from 12ET in the inlays. The printed template and the data below is from the WFret program. The 7th inlay at the neck-body junction is the closest to being "dead-on", at 205.5 mm instead of the theoretical 204.9 mm.

It's hard to completely eliminate optical distortion, this is good as I can make it right now.

Scale Length: 616.0 mm, 9 frets 12ET
Fret Number mm from Nut mm from Fret
=========== ============ =============
1 34.6 34.6
2 67.2 32.6
3 98.0 30.8
4 127.1 29.1
5 154.5 27.4
6 180.4 25.9
7 204.9 24.4
8 227.9 23.1
9 249.7 21.8

jdowning - 7-10-2010 at 05:01 AM

Perfect! Thanks for posting this helpful information Fernandraynaud.
It is interesting to note that the apparent deviation of the fret inlays from 12TET seems to pretty well match that of my lute frets after adjustment for fine tuning.

The image resolution is insufficient for me to read the ruler scale accurately.Would it be possible for you to measure the exact placement of the inlays (to the centre of each inlay) measured from the nut for comparison. The tabulation that you provide is only the calculated 12TET fret placement (that I agree with) but does not quantify the deviation of the inlays from 12TET.

For a vibrating string length of 616 mm the 7th fret on a Pythagorean scale would be 1/3 string length or 205.33 mm which is very close to the actual position of your fret inlay #7. Not that I am saying the inlays represent Pythagorean scaling because - judging from the apparent deviations from 12TET - I would say that they do not.

Could you also measure the neck joint to centre of sound hole distance and sound hole centre to bridge to see if these are also 1/3 string length?
Also can you confirm that the oud was made in Egypt (from the label perhaps) or is it an Egyptian style oud that was made elsewhere in the Middle East. Date?
Thanks again.

fernandraynaud - 7-10-2010 at 12:37 PM

The only thing I know for sure is that the oud has a Gawaret business card glued inside. That to me only says they sold it at some point. Gawaret I understand only recently started making any of their instruments. The info on the card might provide clues, but it's pretty recent, they didn't have web sites in the Mezozoic. Was the shamsa removed just to put a business card inside? Don't know, there's no recent regluing obvious, but it doesn't have to be obvious. It was purchased in Egypt I was told less than 2 years ago, and I can't confirm this, but it sure doesn't "feel" like a two year old oud, I'd guess at least 10 years old, not 40, but not 2, just the particulate in the crevices, the smell, the cracks, the look of the mummy strips in the bowl. Maybe they age fast, poor things, peddlin' their butts on Mohamed Ali street in that heat. Ach, and the dust and the auto exhaust! And it sounds deep and fat, not at all like new wood.

On the enclosed you can estimate the deltas pretty accurately, in reference to the 12ET table, zoom in, but I'll take more exact measurements you ask for later this weekend.

jdowning - 7-10-2010 at 04:07 PM

Thanks Fernandraynaud. The higher resolution image should be good enough as it is probably about as close as it is practical to measure. My readings recorded below.

Oddly enough, the closest match to historical lute fret placement that I can find - for a string length of 616 mm - is that given by the Spanish theorist Juan Bermudo in his 'Declaracion de instrumentos musicales', 1555 - for a vihuela (equivalent to a lute).

-------- Vihuela -------- Oud --------- 12TET------- 1/8 Comma MT

Fret 1 - 35.5 ----------- 36+ --------- 34.6 --------- 35.6
Fret 2 - 67.3 ----------- 66+ --------- 67.2 --------- 66.9
Fret 3 - 96.3 ----------- 97------------ 98.0 --------- 98.7
Fret 4 - 127.3 --------- 127.5 ------- 127.1 --------126.1
Fret 5 - 154 ------------154 --------- 154.5 ---------154.8
Fret 6 - 180.1----------181---------- 180.4 ---------179.2
Fret 7 - 204.5 --------- 205 --------- 204.9 -------- 204.6

So, if my lute had inlays set in the same relative positions as those on your oud, the lute would be pretty much in fine tune if the frets were tied directly over the inlays compared to setting them at 12TET spacing initially.
This would be equivalent to a meantone temperament varient - something like, say, 1/8 comma.
Interesting.

fernandraynaud - 7-14-2010 at 03:39 PM

Jdowning, I have some doubts here. This goes back to my previous issue about frets affecting all notes in a row. I can set my Korg OT120 tuner to various temperaments, from Pythagorean to Meantone to Valotti, a dozen different tunings appropriate for Renaissance and Baroque music, that I can use to tune my harpsichord and clavichord. If I tune my oud's AA course to 440 Hz, and my other courses say in perfect fifths, I can step through various temperaments on the Korg and see if any agree with the inlays on the Egyptian. In other words if I finger on those lines, is there a temperament setting on the tuner that gives me close to zero deviation for all notes on all strings. So far I don't see one. You might get a "hit" say for meantone on one string, if you finger at the inlay lines the notes will agree with the tuner, but then on other strings they will not. The one remaining question is how to tune the open strings relative to each other, but I feel confused as to how lutes are made to conform to say a meantone tuning on all strings.

An advantage of marking strings rather than the fingerboard is we can set intervals separately on each string, rather like the staggered frets I referenced above.

BTW, I have a related question for you. During our Bay Area Oudapalooza dinner here, Michael asked if Maqamat can be freely transposed, and a discussion ensued. It occurred to me later that one reason for "not really" is because the half flats are largely static, for starters in most maqamat where they occur they are B half flat (Iraq) and/or E half flat (Sikah). I realize that there are others and many subtleties. I want to rig a harpsichord with levers so I don't have to do tricks with retuning, and realized that if I could only get B and E half flats, that would cover a vast number of cases. It's these, and it seems only these, that have their own names. Could it be that when the oud had frets, it evolved with a limited number of quarter tone frets, at least one between the first and second chromatics. Has this been discussed and where?

jdowning - 7-15-2010 at 05:21 AM

Both the past Arab and Persian fretting systems for oud and fretting systems for lute are all on historical record. I have plotted some of these to the same relative scale for comparison - posted and discussed on the last page of my "Old Oud - New Project" topic on the Oud Projects Forum, so there is no need to go over the same ground again here.
The early ouds were generally tuned in fourths throughout and only fretted as far as the 'fourth finger fret' (equivalent to the 5th fret in 12TET) and - yes - there were 'quarter tone' (or rather intermediate) frets. For example one of Al-Farabi's systems gives a total of 10 frets up to and including the fourth finger fret with intermediate fret positions between the first, second, third and forth chromatic fret positions (12 TET). However, some frets were only used for some strings and not others. Furthermore some of the frets were not always used by some players. I guess it all depended on what mode was being employed and what a player of the mediaeval time period considered to be acceptably in tune. Like early keyboard music that only sounded well in a few restricted 'keys' ('meantone' tuning)- perhaps the fretted ouds (unlike modern unfretted ouds perhaps?) also were only used for a restricted number of maqamat? I don't know the answer and I am not aware if this question has been subject to scrutiny by researchers.

On the other hand it has been possible to examine the feasibility of the historical lute fretting systems by testing them against lute tablatures of the 16th and
17th C. For example Mark Lindley has recently done just that and concisely records the results in his book "Lutes, Viols and Temperament" (Cambridge University Press). I have not had time to read and absorb all of the material presented by Lindley but his overall conclusion seems to be that lutes were generally (but not always) fretted close to equal temperament - fine adjustments to the fret positions being required because of lack of uniformity of the gut strings, significant diameter differences between the gut (or silk) treble and bass strings, fret height variations, string tension variations when a string is pressed onto the finger boards etc. The lute, of course, is a polyphonic instrument so the fret spacings will be adjusted for the best 'in tune' compromise judged by the player for all tone combinations required to perform the music.
Note also that in the historical perspective lutes and keyboard instruments were not considered to be compatible as far as tuning is concerned.

So, to get back on topic, having determined from the above 3 examples that the finger board inlays are not random but likely represent fairly precisely fret or finger positions (12TET or whatever), the question that interests me is from what tradition and time period does the practice of inlaying an oud finger board in this fashion originate? Is it just that some luthiers in recent history - for some as yet unknown reason - have replicated (rather exactly however!) the finger boards of Western fretted instruments using inlays instead of metal frets. Alternatively does this practice have its origins in a much earlier time period when ouds were fretted?

Good luck with your modified harpsichord experiment fernandraynaud. You might be interested in Cris Forster's book "Musical Mathematics" (on the art and science of acoustic instruments) soon scheduled for release. Excerpts from the book (including analyses of some of the early oud fretting systems as well as 'just intonation' experimental instruments etc.) being freely available at

http://www.chrysalis-foundation.org

fernandraynaud - 7-15-2010 at 05:13 PM

Thanks! I think I see a simple way on the harpsichord to at least give me Ebb and Bbb at will. If It works out I'll post.

An interesting thing is that although a decidedly 1/4 tone half flat, say 50 cents down, sounds awful in chords, a less extreme one can sound good!

jdowning - 12-17-2010 at 06:26 AM

There is an interesting paper by Amine Beyhom and Hamdi Makhlouf, l'Uninversity de la Sorbonne, Paris - "Frettage du Ud ..." - where the authors explore and analyse the historical evidence, for and against, supporting the use of frets on the oud during earlier periods of history.
The basic question is were the frets ('dasatin') 'real' - i.e. gut strands tied on like the frets of a lute - or were they simply reference marks scribed on the fingerboard?
The approximate marking of fret positions on a lute fingerboard would have been a useful reference for the player yet, to my knowledge, no surviving lutes are marked in this fashion. Instead the lutenists of the 16th and 17th C were dependent (to some extent) on fretting instructions found in books of lute tablature.

There is no question that some of the early Arab/Persian theorists were referring to 'real' gut frets - as there is reference not only to the material of the frets but also advice on the quantity of gut required to fret an oud. The venerable Tunisian 'oud arbi' has also retained tied on frets to this day.
I have only had chance to glance over the article and it will take me some time to go through it in detail (it is in French) so I am curious to see if the authors have been able to determine when oud frets became obsolete.
The paper is also a very useful reference with footnotes giving details of source documents etc.

So, are the 'fret' inlays found on some old ouds today a remnant of a tradition that dates back to a much earlier period in the history of the oud - or are the inlays just a copy of, say, Western guitar fret positions and so just intended as a decorative feature by the luthier?

The article can be downloaded at:

http://gfax.ch/literature/

as posted on the topic "Free Music PDF Library" by robertthomasmartin

Brian Prunka - 12-17-2010 at 01:02 PM

Quote: Originally posted by fernandraynaud

BTW, I have a related question for you. During our Bay Area Oudapalooza dinner here, Michael asked if Maqamat can be freely transposed, and a discussion ensued. It occurred to me later that one reason for "not really" is because the half flats are largely static, for starters in most maqamat where they occur they are B half flat (Iraq) and/or E half flat (Sikah). I realize that there are others and many subtleties. I want to rig a harpsichord with levers so I don't have to do tricks with retuning, and realized that if I could only get B and E half flats, that would cover a vast number of cases. It's these, and it seems only these, that have their own names.

A little OT, but this sounds like a cool idea. I would say that if you are going to the trouble, you should include A half flat and F half-sharp, as they are both very common. D half flat is less common, but occurs with some regularity in the repertoire.