A gilt-brass astrolabe signed by Ahmad ibn Abu 'Abdallah al-Qurtubi al-Yamani, North-East Spain, Tudela, dated 737 AH/1336-37 AD

A REMARKABLE 14TH CENTURY ANDALUSI ASTROLABE SHOWING YEMENI INFLUENCE

Several dozen astrolabes are known from al-Andalus, the term we use to denote that part of the Iberian Peninsula under Muslim domination at any time. They date from the tenth to – inevitably – the fifteenth century. Altogether some fifty-odd astrolabes with inscriptions in Arabic, Hebrew or Latin, have all been catalogued in great detail by Dr. Azucena Hernández Pérez of Madrid, a monumental achievement given the diversity and complexity of the instruments and their geographical distribution in museums and private collections around the world (see Pérez 2018).

This gilded brass astrolabe of is signed by Ahmad ibn Abi ʿAbdallah al-Qurtubi al-Yamani in Tudela in 737 Hijra, that is, 1336/37 AD. The maker was previously unknown to us and his production is the only known medieval astrolabe from Tudela, a city divided between three religious communities at that time (see Viguerra 2000). Captured by the Muslims in 813, it developed into as a centre for agricultural and commercial activity. The Muslims remained there after the reconquista as Mudéjars and Moriscos until their expulsion in 1614. When the Muslims expanded out of Arabia in the early ninth century, Yemenis were among them. For example, one of the men charged with laying out the Grand Mosque in Cordoba in the 780s was a certain Hanash al-Sanʿani, from Sana'a (Hanash being a Yemeni name).

The instrument is carefully fashioned from a technical point of view and all letters and numbers are elegantly engraved in an elegant Andalusi Kufic. All numbers are expressed in the Western abjad convention (with صاد, sad for 60, rather than س, sin). Since an integral component of this new astrolabe, the alidade for measuring celestial altitudes, is missing, as well as one of the two plates, notably, one for Tudela, the instrument is non-functional in its present state.

The instrument is gilt, although this is most probably a later addition. Very few medieval or Renaissance instruments are gilt – see, for example, one very medieval-looking Renaissance astrolabe in the Museo Nazionale del Bargello in Florence, which is unusual but not in fact in any way dubious once it is realised that it was inspired by an Andalusi astrolabe (King 2011, p.385, fig.19 and p.398, no.4507).

There is clear evidence beyond the maker’s epithet al-Yamani to confirm that the maker or his family or his ancestors hailed originally from the Yemen. There is a Yemeni astrolabe in the Metropolitan Museum of Art, New York, inv. no.91.1.535a–h, made and signed by the Yemeni Rasulid prince al-Ashraf in 695 AH/1295 AD (King 2005, pp.615-657, XIVa). This is a fine example of a thousand-year tradition of serious astronomy in the Yemen (see King 1983). The astrolabe under discussion is a fine example of the seven-hundred-year tradition of astronomy in al-Andalus (see Samsó 2000).

On the back of al-Ashraf's astrolabe there is an astrological scale in which symbols are used for the sun, moon and five naked-eye planets. Such symbols, Greek in origin, are first attested in the Islamic tradition on the spectacular astrolabe by the astronomer al-Khujandi in Baghdad in 374 AH/984-5 AD (now in the Museum of Islamic Art, Doha), but also, what is possibly more significant, on the astrolabe made by the Yemeni Sultan al-Ashraf mentioned above, on which all the scales on the back are replete with such symbols (for detailed descriptions of al-Khujandi’s astrolabe and all others from between 750 and 1100 AD see King 2005, vol.2 pp.403-544, XIIIb-c: esp. pp.502-517, and on the quatrefoil: p.510).

If the maker of the new astrolabe was originally of Yemeni extraction, he had certainly mastered ornamental Andalusi Kufic. It is clear from his epithet al-Qurtubi that he considered Cordoba to be his home or his most recent residence.

The rim of the mater bears a scale clockwise from zero to 360°, divided and labelled for each 6°: و - يب - ... - شص , 6 - 12 - ... - 360, subdivided for each single degree. The rim and the throne are of one piece and are soldered onto the mater (which alone is not gilt).

The throne is low and pierced with three holes; it is in the tradition of the prolific astrolabist al-Khama’iri of Seville, c.1200. Perhaps the most remarkable feature of this Tudela astrolabe is the pair of rosettes or hexafoils decorating the rete, not attested on any other Islamic (or European) astrolabe. For further information on quatrefoils on Western Islamic, Medieval European and Inscribed with Hebrew Inscriptions and Renaissance astrolabes, see D. King, 'The quatrefoil as decoration on astrolabe retes', 2005, pp.963-991. It is rare that one can trace a decorative feature on artefacts over centuries and across continents.

The astrolabe of al-Khujandi features a single quatrefoil in the upper part of the ecliptic ring, surely of Byzantine inspiration. Note, however, that in the Latin world, significant quatrefoil decoration was found, for example, on a Catalan astrolabe from c.1300 now in the society of Antiquaries in London, and a yet more spectacular English example also from c.1300, that is, 'The Sloane astrolabe' now in the British Museum, inv. no.SLMathInstr.54 (Gunther 1932, pp.307 and 462 and King 2005 'Quatrefoil', pp.982 and 985).

In addition, these rosettes are connected to a circular frame at the bottom of the rete by two rectilinear segments, which replace the standard circular equatorial bar. This feature is not known from any other Islamic or European astrolabe, but is neither a good idea nor a bad one.

On the rete or star-map, the ecliptic scale is divided into 12 divisions for the zodiacal signs: Aries – Taurus – Gemini – Cancer – Leo – Virgo – Libra – Scorpio – Sagittarius – Pisces. (The word العقرب is misspelled العقراب).

The ecliptic scale is divided and labelled for each 6° within each sign (36-24-18-12-6).

The rete itself is unusual for an Andalusi one, indeed it bears no relation to surviving earlier or later Andalusi retes. There are four knobs for turning the rete over the plate beneath. The horizontal bar is counter-changed twice on each side of the central disc, a purely decorative feature. The star-pointers are mainly short and not decorated, and they resemble those on Eastern Islamic astrolabes more than those on Western Islamic ones. The pointers have smooth cavities on their bases, perhaps for a silver inlay, as on some Andalusi astrolabes, which have flat-surface silver inlay on the pointers.

The stars represented and their modern designations are as follows:

1st quadrant (counter-clockwise from the left):

batn qaytus - ghul - al-dabaran - ʿayyuq - rijl al-jawza - mankib al-jawza

ζ Ceti - β persei - α Tauri - α Aurigae - β Orionis - α Orionis

2nd quadrant:

al-ʿabur - al-ghumaysa - muqaddαm al-dhiraʿan - rijl al-dubb - ra’s al-shujaʿ - rijluhu - janah al-ghurab

α Canis maioris - α Canis minoris - α Cephei - λ Ursae maioris - α Hydrae - μ Ursae maioris - γ Corvi

3rd quadrant:

al-qa’id - al-aʿzal - al-ramih - al-hayya - fakka - qalb al-ʿaqrab - al-hawi

η Ursae maioris - α Virginis - α Bootis - α Serpentis - α Coronae Borealis - α Scorpionis - α Ophiuchi

4th quadrant:

waqiʿ - dhanab al-jady - al-ta’ir - mankib al-faras - khadib - dhanab qaytus

a Lyrae - d Capricorni - a Aquilae - b Pegasi - b Cassiopeiae - b Ceti

This is a Western Islamic selection of star-names rather than an Eastern one, the Western names al-ʿabur and al-ghumaysa being representative of the former (as opposed to al-shiʿra al-yamaniya and al-shiʿra al-sha’miya in the latter). (One could argue that the name of the star muqaddam al-dhiraʿan with a nominative dual has been misspelled but one could equally well argue that it is acceptable in technical Arabic.)

The two decorative hexafoils on the rete are entirely unexpected, as are the triple semi-circular arcs connecting them to the horizontal axis and the single semi-circular arc joining the outer frame to the ecliptic ring. It is known that the quatrefoil was used as decoration on Eastern Islamic astrolabe retes from the tenth century onwards, as on the spectacular Baghdad astrolabe of al-Khujandi, and probably earlier because the original inspiration is Byzantine. It was used on astrolabes by Jewish and Christian workmen in the Iberian Peninsula, more often than on Islamic pieces.

The distinctive and substantial circular frame at the bottom of the rete calls to mind Andalusi craftsmanship and initiative, for example, the partial circle in this position on the astrolabe of Muhammad al-Sabban, made in Guadalajara in 1081/82 AD (History of Science Museum, Oxford, inv. no.52473, see Pérez 2018, pp.111-120). It was a necessary component because the two arms attached to the left and right of it were otherwise connected only to the horizontal bar. There is a dummy pointer without a name on the left-hand lower bar, just next to the left-hand hexafoil, which looks as though it has been attached to the ecliptic ring by a twisted strand of metal solder. On other Andalusi astrolabes we sometimes find mihrab-shaped arches offering such support.

The back bears three circular scales in addition to four altitude scale around the outer rim. Each is divided and labelled for each 6° from zero, 6°, 12°, ... , 90° in each of the four quadrants, all subdivided for each single degree. Immediately inside this is a solar longitude scale, with twelve divisions for the zodiacal signs, each divided and labelled for each 6°, subdivided for each single degree. The names of the signs are:

الحمل - الثور - الجوزا - السرطان - الاسد - السنبلة - الميزان - العقراب - القوس - الجدي - الدلو - السمكة

with the same orthographical variant العقراب for العقرب , and the old-fashioned al-samaka for Aquarius. The former is found on the rete. The latter is found on the majority of eighth, ninth and tenth-century astrolabes from the Islamic East but seldom thereafter. Separated from these 360°-scales is a calendrical scale showing the solar months of the Western calendar, divided for each month and the appropriate number of days, mainly 30, but also 28 and 31. The names of the months are:

ينير - فبرير - مارس - ابريل - ماية - يونيه - يوليه - اغشت - شتنبر - اكتوبر - نونبر - دجنبر

Inside this is the scale for the astrological 'terms' (hudud) of the zodiacal signs, five areas of unequal size for each sign, whose length in degrees is indicated by the number below the symbol of the planet associated with that term. The planetary symbols are Greek in origin and were used in medieval Europe (see Gettings 1981). Their appearance here should not be construed as having any occult or magical significance: they are used only because they are convenient given the lack of space. al-Ashraf’s astrolabe has three astrological scales with such symbols. They are not known from Andalusi astrolabes, so we must look back to the beautiful astrolabe of al-Khujandi (Baghdad, 984 AD) and the remarkable astrolabe of the Yemeni al-Ashraf (Yemen, 1295 AD), where the same symbols occur. It is highly probable that al-Qurtubi encountered them on a Yemeni astrolabe.

The horizontal diameter has not been engraved. Above where it would be there is a rectangular frame around which is engraved the principal inscription:

صنعه احمد ابن ابو عبد الله القرطبي اليماني بتطيلة سنة ذ لز هجرية

'Made by Ahmad ibn ʿAbdallah al-Qurtubi al-Yamani

in Tudela (Tutila) in the year 737 Hijra [1336/37 AD].'

Below where the horizontal diameter would be, there is a double shadow square for finding tangents and cotangents and their inverses. The double horizontal scales are each labelled المبسوط, al-mabsut, 'horizontal (shadow)' and the two scales are divided and labelled for each three units: ج - و - ط يب , 3 - 6 - 9 - 12. The two vertical scales are both labelled المنكوس , al-mankus, 'vertical (shadow)', and are divided in the same way.

The single plate bears a set of standard astrolabic markings for a specific latitude on one side and a set of universal markings for all latitudes on the other side. We first consider the former, which are stated to be for قرطبة لح ل , 'Cordóba, (latitude) 38°30'. It seems that al-Qurtubi made his astrolabe in Tudela with Cordóba in mind. Without a dedication and no indication of a commission, we cannot say more. Cordóba, with this latitude, featured on the plates of virtually all Andalusi astrolabes (see King 2005, vol.1, XVI: 915-962, for an analysis of geographical data on early Islamic astrolabes).

Yet the inclusion of a set of universal markings on the other side of this plate could also have served the latitude of Tudela, which is about 43° (most Andalusi astrolabe plates include Saragossa, but none have Tudela). In addition to the three base circles serving the winter solstice, the equinoxes and summer solstice, the meridian is shown as the vertical diameter together with the azimuth circles marked and labelled for each 6° around the horizon, starting at the meridian. The horizon is shown together with altitude circles marked and labelled for each 6° above the horizon up to the zenith, marked as 90°. At the ends of the horizon are the east and west points, labelled المشرق and المغرب , al-mashriq and al-maghrib. Below the horizon are the curves for the seasonal hours, labelled ا - ب - ج - ... - يب , 1, 2, ... ,12. The meridian corresponds to the sixth hour, that is, midday, and is labelled الزوال , al-zawal. To the left of the meridian there are two dotted curves running between the seventh and eighth seasonal hours and between the ninth and tenth seasonal hours: these are marked الظهر and العصر , for al-zuhr and al-ʿasr, the two afternoon prayers (on the prayer-times in Islam see 'On the times of Muslim prayer', in King 2004, vol.1, IV: pp.529-622, and also the article 'Mikat' (astronomical timekeeping and times of prayer) in Encyclopedia of Islam, VII, 1990, pp.27-32).

The zuhr prayer was defined differently in al-Andalus from Eastern Islam, where it was usually defined as midday or very shortly thereafter. The dotted curves are crudely engraved but they seem to be original not least because they are standard on Andalusi astrolabes but also because the two names of the prayers appear to be in identical script to the term al-zawal for midday. There is a notch at the top of the plate to fit into a small hole at the top of the mater and secure the plate.

There is room between the mater and the rete for one more plate. This surely had a set of markings for the latitude of Tudela on one side and a set for another latitude on the other. Most surviving early Andalusi astrolabes have several localities marked on each plate, but Tudela is not attested on any of these (on the geographical data on astrolabe plates see King 2005, vol.2, XVI: pp.915-962, esp. pp.951-7).

On the other side of the surviving plate is a set of universal markings of the kind associated with the late-twelfth-century Granada astronomer Ibn Baso (see Calvo Labarta 1996). They featured on most Andalusî astrolabes after his time, and likewise on Maghribi astrolabes until the nineteenth century. The markings are universal in that they can be used in all latitudes. They could, of course, be used for Tudela. The markings consist of two sets of orthogonal circular arcs, each shown with the appropriate arguments for each 6°. The nodes on the left and right are labelled المشرق , al-mashriq, 'east', and المغرب , al-maghrib, 'west'. The markings are appropriately labelled لجميع العروض , li-jamiʿ al-ʿurud, 'for all latitudes'.

The alidade or sighting device for measuring celestial altitudes is missing. What is still in position are an eight-sided nut and a bolt that fits into it but has no thread. The bolt has a head like the knobs on the rete. The surface of the back underneath the nut gives the impression of its having been used, in spite of the fact that without an alidade the ensemble is non-functional.

Appendix: Notes on the three surviving Yemeni astrolabes

The Yemeni astrolabe by the Rasulid ruler al-Ashraf in the Metropolitan Museum of Art in New York (inv. no.91.1.535, diam. 15.5cm) has been studied in detail in light of the thousand-year long tradition of astronomy in the Yemen. A detailed description of the instrument, together with a presentation of remarks by the Prince’s teachers about six astrolabes that he made (one of which is the New York astrolabe), is in King 2005, vol.2, pp.615-646. Further information is on the website: www.metmuseum.org/art/collection/search/444408.

A second Yemeni astrolabe is preserved in the Institut du Monde Arabe in Paris (inv. no.AI.86.15, diameter 132mm). Six sets of markings for Yemen and Hijaz. It is briefly described in the handlist of the collection prepared by Jeanne Mouliérac in 1989. There is no mention of this piece on the Institute’s website.

Jeanne Mouliérac, Paris IMA Catalogue = La collection Marcel Destombes (1989), pp.89-90 (no.5). A detailed description is in King 2005, vol.2, 2005, pp.649-653.

A third Rasulid Yemeni astrolabe is preserved in the Collection of Historical Scientific Instruments at Harvard University. On their website it is labelled "Persian circa 1590”. For a description, see King 2005, vol.2, pp.649-653. See also: http://waywiser.fas.harvard.edu/search/astrolabe.

For the preparation of this report, Sotheby’s gratefully acknowledges the assistance of Prof. David A. King of the Goethe University in Frankfurt.

Bibliography:

Labarta 1996

Emilia Calvo Labarta, 'Ibn Baso’s astrolabe (plate) in the Maghrib and East', in Josep Casulleras and Julio Samsó, eds., From Baghdad to Barcelona. Studies in the Islamic Exact Sciences in Honour of Prof. Juan Vernet, 2 vols., Barcelona: Instituto 'Millás Vallicrosa' de Historia de la Ciencia Árabe, 1996, II, pp.755-767, at www.academia.edu/34702172/.

Encyclopaedia of Islam, 2nd edn. (available in English and French), 13 vols., Leiden: E.J. Brill, 1960-80.

Pérez 2018

Azucena Hernández Pérez, Astrolabios en al-Andalus y los reinos medievales hispanos, Madrid: La Ergástula, 2018. (detailed scientific and art-historical descriptions of some 50 astrolabes).

Gettings 1981

Fred Gettings, Dictionary of occult, hermetic and alchemical signs, London & Boston & Henley: Routledge & Kegan Paul, 1981.

Gunther 1976

Robert T. Gunther, The Astrolabes of the World, 2 vols., Oxford: Oxford University Press, 1932, repr. in 1 vol., London: The Holland Press, 1976.

Mouliérac 1989

Jeanne Mouliérac, 'La Collection Marcel Destombes', in Astrolabica No. 5 – Études 1987-1989, Paris: Institut du Monde Arabe & Société internationale de l’Astrolabe, 1989, pp.77-126.

King 1998

David A. King, 'Notes on Yemeni astronomy in the Rasulid period', an essay review of Daniel Martin  Varisco and G. Rex Smith, eds., The Manuscript of al-Malik al-Afdal (d.1377) – A Medieval Arabic anthology from the Yemen, [London]: E.J.W. Gibb Memorial Trust, 1998, in Yemen Update 44 (2002), accessible at www.academia.edu/34703707/.

King 1985

David A. King, 'An astrolabe made by the Yemeni Sultan al-Ashraf' (1985), in idem, In Synchrony with the Heavens, vol.2 (2005), XIVa: pp.615-657 (includes descriptions of two more Rasulid astrolabes).

King 1983

David A. King, Mathematical astronomy in medieval Yemen: A bio-bibliographical survey, (Publications of the American Research Center in Egypt), Malibu CA: Undena, 1983, at www.academia.edu/34693220/.

King 2004-5, vols 1&2

David A. King, In Synchrony with the Heavens – Studies in Astronomical Timekeeping and Instrumentation in Islamic Civilization, (Islamic Philosophy, Theology and Science – Texts and Studies, vol. LV), vol.1: The Call of the Muezzin, & vol. 2: Instruments of Mass Calculation, Leiden & Boston: Brill, 2004-05, available at www.academia.edu/43168384/ & www.academia.edu/37189917/.)

King 2005

King, 'The quatrefoil as decoration on astrolabe retes', in idem, In Synchrony with the Heavens, vol.2 (2005), XVII: pp.963-991.

King 2004

David A. King, 'On the times of Muslim prayer', in idem, In Synchrony with the Heavens, vol.1 (2004), IV: pp.529-622, at www.academia.edu/43168384/.

King 2005

David A. King, 'An astrolabe made by the Yemeni Sultan al-Ashraf' (1985), in idem, In Synchrony with the Heavens, vol.2 (2005), XIVa: pp.615-657 (includes descriptions of two more Rasulid astrolabes).

King 2018

David A. King, 'The Astrolabe – What it is and what it is not' (2018), at www.academia.edu/37613545/.

King 2011

David A. King, Astrolabes from medieval Europe, Farnham & Burlington VT: Ashgate–Variorum, 2011 (on the astrolabe in its medieval context and on the misinformation of the instrument on the internet).

Samsó 2020

Julio Samsó, On Both Sides of the Straits of Gibraltar – Studies in the History of Medieval Astronomy in the Iberian Peninsula and the Maghrib, Leiden & Boston: Brill, 2020 (the most important work on the history of astronomy in al-Andalus).

Viguera 2000

Maria J. Viguera, 'Tutila' in Encyclopaedia of Islam, 2nd edn. (available in English and French), 13 vols., Leiden: E.J. Brill, 1960-80, X, 2000, pp.752-3.

For an introduction to the astrolabe see John North’s splendid article:

www.scientificamerican.com/article/the-astrolabe/.

On the use of the astrolabe by a competent non-specialist, Tom Wujec, see:

www.ted.com/talks/tom_wujec_demos_the_13th_century_astrolabe.