epigraphy in Arabic by dov51579


									 On editing epigraphyin Arabic script
 A suggested torm ot editorial presentationx
  b1 F.Th. Dilkemd

 The Muslim world is very rich in epitaphsand monu-                                                                                    survive. the major representatives the former cate-
 mental and other inscriptions,which contribute greatly                                                                                gory being long texts of an expository or literary
 to our knowledgeof the historiesof individuals,dynas-                                                                                 character,and those of the latter being short texts of a
 ties. tou'ns. states,monuments and art. About a cen-                                                                                  documentary character (papyri. inscriptions). Owing
 tury ago major attempts first began to be made to                                                                                     to thesedifferences editorial treatment of the texts
 collect thesetexts and to make them accessible the to                                                                                 of thesetwo categories   also differs.While much of the
 scholarly world. Since that time the texts of probably                                                                                task of the editor of a text of the former category
 several thousands of such inscriptions have been                                                                                      consistsessentially collation. the editor of a text of
 published.                                                                                                                            the latter categorywill necessarily engagemore specific-
    While students of Muslim epigraphy have so far                                                                                     ally in researchon matters relating to palaeography,
 concentratedon rendering accessible many texts as
                                         as                                                                                            terminology, histor,v etc. And whereas the former
 possible together with the details which the reader                                                                                   editor can in his edition justifiably limit himself to
 mav need for a correct understanding       (translations.                                                                             reproducing. besideswhat he beliel'esto be the text
commentaries.      photographs. indexes. comparative   stu-                                                                            clo -,t to the original,only the most important variant
 dies etc.). little thought seems have been spent on
                                  to                                                                                                   deti,.,s the variouscopiesused,the latter editor can,
 how actually to present these materials. Yet a rigid                                                                                  and indeed, in order to serve the reader best, must,
 form of editorial presentationcan help very much to                                                                                   reproducea large variety of detail of the unique copy
 facilitate the consultation of an edition, to prevent                                                                                which he is editing.
 unnecessary    diffusenessand to provide the editor with                                                                                When addressingthe issue of how to publish the
a usefulcheck on his own conclusions.                                                                                                 texts of inscriptions and papyri, one must consider
    Before enteringinto a discussion the techniqueof
                                      of                                                                                              what questions the critical reader may ask of an
editingepigraphyin Arabic script.it is appropriateto                                                                                  edition. These questions are essentially:what text is
define in general the editor's task vis-à-visthe reader                                                                               available?what portions if any are lost? which of
and the demandsmade by this kind of material.and,                                                                                     thesehave apparentlybeenlost accidentallyand which
in order to prevent duplication and the rise of a                                                                                     through wilful deletion? what may have been the
uselessvariety of editorial systems, to see what                                                                                      content of such lost portions'l what text, though avail-
thought studentsof epigraphy in other scriptshave so                                                                                  able, cannot be read? of which among the available
far given to this subject and what conventions they                                                                                   letters is the reading doubtful? how are any abbrevia-
use.                                                                                                                                  tions or contractions available in the text to be filled
   The verb 'to edit' means 'to prepare for publica-                                                                                  out? what errors does the available text contain?
tion'. From the editor's point of view this means                                                                                     which of these errors can be assumedto be due to
'while                             justice to the original
        doing as much as possible                                                                                                     contamination of whatever kind and which to mere
text. to present it in a manner as convenient as                                                                                      inadvertenceon the part of the scribe?what alterna-
possiblefor the reader'1.With modern texts of living                                                                                  tive readings deserving serious consideration have
authors this means that with the approval of the                                                                                      been suggestedby others? what is known of the
author the editor may alter the wording or the arrange-                                                                               history of the document itself? what are its physicial
ment if this helps to make the meaningmore apparent.                                                                                  characteristics?  what is its location? and how is it to
ln the caseof older texts, however. where the creator                                                                                 be interpreted?
of the work can no longer be consulted,the editor has                                                                                    It was only towards the end of the nineteenth
no such liberty but must render the text faithfully.                                                                                  century that editors of inscriptionsand papyri came to
   Merely from the point of view of editorial handling,                                                                               feel that they could only servescholarshipin the true
two major categories can be distinguished among                                                                                       senseof the word by presenting their editions in a
theseolder texts, viz. that of texts surviving in several                                                                             form answeringall of thesequestions;until that time
later copies only and that of texts whose originals                                                                                   they had mostly been content with presentingfacsimi-

M a n u s c r i p t so f t h e M i d d l e E a s t ó ( 1 9 9 2 )   ( ' T e r L u g t P r e s s .D o n k e r s r e e g1 9 . 2 3 1 2 H A L e i d e n . N e t h e r l a n d s .1 9 9 , 1
                                                                                                                                                                                        ISSN 0920-0401
130                                    MANUSCRIPTS OF THE MIDDLE [,AST 6 íI99])

les, or something standing betweena facsimile and a           The stressplaced by the Leiden Committee on the
text set from type, provided with scanty supplemen- need for uniformity is entirely justified. Considering
tary details. Severalmethods of critical editing came the numerous particularitieswhich the editor must
to be developedconcurrently and soon the need was indicate in order to provide the reader with all the
felt among epigraphistsand papyrologistsfor unifor- detailswhich he may want to know. any system            will of
mity. The matter came to be discussedincreasingly. necessitybe complex. In order to spare the reader
and in 1928the Royal Academy of Belgium presented from having to master a new system whenever he
to the Union Academique Internationale a plan for consultsanother edition, this uniformity is essential.
the unification of the methods followed and of the Deviation from the Leiden System is therefore to be
conventional signs used in scholarly text editions, limited preferably to those instanceswhere. for what-
including editions of inscriptions and papyri. The ever reason.the systemdoes not suit the Arabic script:
Union took action and in 1932 published a brochure and deviation from Grohmann's adapted version is
entitled Emploi des signes critiques. Disposition de justiÍied only insofar as such a deviation is a patent
l'upparat duns les éditions sarantes de textes grecs eÍ improvement refinement.
latin.s.Conseilset recommandations2.      Meanwhile the       The questionswhich the critical reader may ask of
question had also been discussed the papyrological an edition have already been listed above. We shall
section of the l8th international congressof orienta- now discussthesein further detail with soecificrefer-
lists.held in Leidenin 1931.After having taken notice enceto inscriptions Arabic script:l
of various views. including what must have been a
                                                           (1) Ertunt tc.rr.As regardsthe extant text of a docu-
preliminary version or proofs of the brochure of the
                                                           ment in Arabic script.one might publishit in translite-
Union. the Section on 1l Septemberof that year
                                                           ration. u'hichmay prove easier print. However.this
unanimously acceptedthe text of a resoiution prepared
                                                           ma,vocciisionall.v   necessitateinterpretationof a kind
by a committee of three members of the Section.
                                                           which at this stageis undesirable.   and will complicate
entitled Essai d'unific:ation des methode,semployée,s
                                                           comparison with the original document. Moreover.
clonsle.séclitionsde pap,vrus3. This document proposed
                                                           the reader may find it easier to read the text in the
a well-considered   systemof conventionsfor editors of
                                                           original script. Having said this I think the reader is
papyri which, though new, followed as closely as
                                                           served best if the edition is not a facsimile but is
possiblethe systemsalready in use, in order to avoid
                                                           printed from regular Arabic type: in instanceswhere
undesirablediscontinuity. In the same resolution the
                                                           the writing in the original document is very intricate
Section also stressedthe desirability of as great as
                                                           and the reading is disputable,the editor may in the
possiblea uniformity in the principlesfollowed in the
                                                           criticalapparatus   present drawing of the contoursof
editing of all unique documents.     not only papyri but
                                                           the letter or letters concerned as they appear in the
also such others as Greek and Latin inscriptions.      (It
                                                           original document.
would seemthat the existence papyri and inscrip-
                                                              In inscriptions Arabic script diacriticalsignsare
tions in other languages   and the need for their publica-
                                                           not alwayscompletelyavailable.while such auxiliary
tion. too. did not occur to the Committee.)TheUnion
                                                           signs as vowels tend to be almost completely lacking.
in its turn responded 1938by publishinga revised
                                                           While on occasion this absenceis due to apparent
edition of its brochurea. which included comments
                                                           inadvertence on the part of the scribe, often it is
and suggestions    for a slight revision of some of the
                                                           evident that these signs have been omitted inten-
proposalsof the Leiden Committee.However, this did
                                                           tionally, as a calligraphiclicence.The question of how
not prevent the new systemfrom coming to be known
                                                           the editor is here best to serve the various interests
as the Leiden System or. in German. Ihe Letdener
                                                           involved is difficult to answer: while the epigraphist
Klamntersyslerr.  Sincethat time various scholarshave
                                                           may prefer an edition presentingexclusivelythe signs
published their comments on the theory of certain
                                                           available in the original document, the historian may
conventions among those proposed by the Leiden
                                                           prefer an edition that makes reading as easy as pos-
Committee, andlor suggestedchanges or refinements.
                                                           sible. At the same time both typesettingof the auxi-
Some of these suggestions    have been adopted, others
                                                           liary orthographic signs proper to the Arabic script
have not. The most comprehensiveattempt of this
                                                           and deletion of the regular diacritical dots add signifi-
kind is the brochure by the specialistof Greek epi-
                                                           cantly to the cost of printing. Last but not least. one
graphy at Harvard University. Sterling Dow, entitled
                                                           may wonder whether these signs with their mostly
Conventionsin Editing. A SuggestedReJ'ormulatíon        of
                                                           secondary importance for the understanding of the
the Leiden St'stem.which was publishedin 1969.sThe
                                                           text deservethat amount of the editor's attention. On
scholar who introduced the systemin the field of the
                                                           the basis of theseconsiderationsI suggestthe follow-
study of documentsin Arabic script is probably Adolf
                                                           ing compromise:
Grohmann. Specifically for the use of students of
Arabic papyrologyhe published, 19526 in       and several (a) To supply tacitly diacritical dots whose lack is not
times subsequently,                                                                        (in
                      summariesof the systemadapted due to apparentinadvertence which event an edito-
for use in editions in Arabic script.                      rial substitution is required (seebelow)).
                                  F . T H . D I J K E M A . O N E D I T I N G E P I G R A P H YI N A R A B I C S C R I P T          t3l

(b) In editions of texts in languages other than Arabic,                       texts. With speciÍic reference to verse inscriptions.
to supply tacitly any of the lacking supplementary                             records of thesemay be found in díw,ans.      With some
diacritical signsusedin the writing of those languages,                        critical caution the modern editor can safely supply
unlesstheir absence the original document is appa-
                       in                                                      from such records portions now lost. For the critical
rently due to inadvertence(in which event, again, an                           reader it is essential to know. however. that these
editorial substitution is required).                                           portions, though lost, are not conjecturesbut have
(c) To omit all other auxiliary orthographic signs,but                         been borrowed from existing records. Martin West,
in Arabic texts to retain consistently or to supply                            the author of the most authoritative modern hand-
tacitly some, viz., following the rules drawn up by                            book of textual criticismll. therefore stressesthe
Regis Blachèreand Jean Sauvagetfor the contributors                            appropriateness the use (the existence which had
                                                                                                of                        of
to two French seriesof Arabic text editionss..   hamza.                        already been noted in the brochure of the Union) of
shadda and madda with the exception of shadda over                             only the bottom half of the square bracket, the halving
'solar'                                                                        indicating the fact that in supplying the text the editor
        letters following the article; vowels and jazma
wherever their absencemight cause ambiguity; and                               stands on firmer ground than if he had made a mere
the two subscript diacritical dots of final ya'used to                         conjecture(in which event he would have usedthe full
distinguishthat letter from alif maqsura.In editions of                        squarebracket)(seesampleNo. 2). Dow's recommend-
epigraphic texts in other languages the same rules                             ation to underline such portions instead,seemsto me
might mutatis mutandisbe applied.                                              lesslogicai.
   It is the practice of Max van Berchem and his
                                                                               (4) Doubt/ul readings.  The reading of individual avail-
lollowers to present in editions not the full texts of
                                                                               able letters may be doubtful owing either to physical
koranic citations as theseappear in the original docu-
                                                                               damage or to lack of accuracy on the part of the
ments but only the references those citations, and
                                                                               scribe. In very many instances,however, the reading
not the full blessingsas they appear in the original
                                                                               of such letters wiil be evident from the context. Yet
documentsafter the names of God, Mohammed. etc.
                                                                               strictly speakingthe readingsof such lettersare conjec-
but only the conventional contractions of these.It is
                                                                               tures. Since these particular letters are not lost or at
true that this method savesspace,but it may deny us
                                                                               any rate not entirely lost, it would be incorrect to
possibly interesting variant readings and should, I
                                                                              place them within square brackets. In classicalepi-
think. not be followed. As regards punctuation, I
                                                                              graphy the fact that the reading of such a letter is.
think the readerwill be servedbest if in prose texts the
                                                                               strictly speaking,a conjecture,is indicated by printing
editor suppliesconsistentlyat any rate full stops and
                                                                               a dot underneaththat letter. In the Latin and Greek
hyphens.and perhapsalso colons.e
                                                                               scripts this critical sign is entirely appropriate, as it
(2) Text lost by accident.Portions of text apparently                          attracts the readersattention and yet causes incon-
lost by accidentare marked in the edition by a square                          veniencewhile reading. Unfortunately, in editions of
bracket on each side (seesampleNo. l). Any conjectural                         our texts the presence the diacritical dots proper to
restorations of such missing text are placed between                           the Arabic script make the further dot undesirable aas
these brackets. In caseswhere the extreme end of a                             critical sign. Grohmann proposes the use of small
line of text is missing and the length of the portion                          superscriptquery marks; I myself have in my edition
missing cannot or not approximately be determined,                             of the Ottoman historicalinscriptionsin Edirnel2 used
the bracket that comes next to the extreme available                           overlining. Superscript query marks seem to me to
letter may alone be placed and the other left out.                            distract the eye too much from the text proper, espe-
However. the use not of a pair of brackets but of a                           cially if they are numerous, and what is more, they
single bracket alone may causeconfusion if preceding                          may suggestthat the reading not only of the letter or
or following lines also contain bracketsor if the editor                      letters concernedbut of the entire word is doubtful.
makes a conjecture(in which event the readermay fail                          Moreover. most printers will find it hard to set super-
to notice the single bracket and consequently   the fact                      script query marks. It is true that many printers will
that a conjecture is being made). This method must                            find it hard to set the superscriptbar too, but if need
thereforebe applied only in those rare instances  where                       be they can quite easilyincorporate this sign by hand.
there is really no way at all for the editor to define                        The superscriptbar has the additional advantage of
even approximatelythe length of the portion missing.                          being familiar to readersof manuscriptsin the Arabic
                                                                              script as a means to make certain portions of text
(3) Borroyringsfromexistingrecords.   The virtual founder                     stand out. Provided that care is taken to avoid pos-
of the study of Arabic epigraphy, Max van Berchem,                            sible confusion with /'atha and medda by placing the
complained as early as 1892 about the rapid decay of                          superscript  bar one registerhigher than the superscript
inscriptionsin the Muslim world.10 Fortunately,inter-                         auxiliary signs of the Arabic script, it will serve its
est in inscriptions as auxiliary sources of historical                        purpose well (seesampleNo. 3).
information appears to have existed already at an                                With reference the marking of doubtful readings
early date, e.g. in Turkey, with the result that all kinds                    it may be useful to add that among inexperienced
of Ottoman chronicles include records of inscription                          editors there is a tendency towards also so markins
t -)/,                                 M A N U S C R I P T O F T H E N { I D D L EI A S T 6 ( I 9 9 ] )

damaged letters whose readingseven in isolation are                     edition). Patent orthographic errors apparently due
beyond doubt. Van Groningenl3 and others have                           solely to inadvertence.  however,shouid for the conve-
stressedthe uselessness this mode of procedure.
                           of                                           nience of the reader be marked. The Leiden System
What intereststhe reader in the first place is what text                advises the editor to place between angle brackets
is legible.As long as the reading of an individual letter               letterswhich he has insertedto supply lettersor words
is beyond doubt in spite of any damage, that letter                     which he considers have beenintendedto be inscnbed
ought to be printed without any crittcal sign.                          but to have beenomitted by error (seesampleNo. 5),
   With respectto the marking of letters of doubtful                    and between braces letters or words which he consi-
reading it must finally be noted that in certain instan-                ders not to have been intended but which have been
ces the Arabic script unfortunately lends itself ill to                 inscribedby error (seesampleNo. 6), and this conven-
such markings, as lettersmay appear not only next to                                                                 If
                                                                        tion has now been generallyaccepted.ls an editor,
one another but also underneathone another (e.g. in                     suspectingone or more entire words to have been
the common ligature lam-mrm).The only solution in                       skipped.ló wishesto make a conjecture.he placesthat
such instances  seems be to have such lettersprinted
                       to                                               conjecture   between  anglebrackets;if in such an event
as if there existedno such ligatures.                                   he wishes to make no conjecturehe inserts the angle
                                                                        bracketsand betweenthem leavesappropriatespace.
(5) Intentionol erasures.  Areas where the scribe or a                  The third kind of editorialcorrection.   the substitution,
later hand have apparently deleted a portion of text                    has often beenplaced betweenangle bracketstoo. The
intentionally are marked in the edition by a double                     use of a singlecritical sign for two different purposes
squarebracket on each side (seesampleNo. 3). Erased                     was found admissiblein that the reader was felt to
letters that can still be read with some certainty (for                 notice the differencein another way: in the caseof a
instancebecausethe contours are still visible) can be                   substitutionthe critical apparatuswouid necessarily
printed betweenthosebrackets;rf the readingis doubt-                    includean entry showingthe word precisely read by
ful, overliningmust be added;conjectures text once
                                             of                         the editor in the original document.while in the event
available there must be printed betweensingle square                    of an editorialaddition the readerwould Íind no such
brackets within double brackets; portions of text                       entr.vin the critical apparatus.  Horvever.  the inconve-
which apparentlyafter the erasurehave been inscribed                    niencecaused      to the reader b1' this cumbersome
in an erased area are printed between the double                        method was recognized an earh' date. Bidez and
                                   'SecondText' printed                 Drachmann,op. t'it.. p. 20. have proposedthe useof a
square bracketswith the words
from small type overhead (if the editor is somehow                      separatecritical sign for marking editorial substitu-
capableof supplying also the first. erasedtext of that                  tions. viz. the top half of the squarebracket.While the
area he puts that text between the double square                         idea of the use of a separate   sign is a good one. the
                           'First Text' overhead.and the                 particular one suggested   has not generallybeen accep-
bracketswith the words
                                           'Second Text'
second text on top. with the words                                       ted. perhaps becauseit was felt to be too inconspi-
 overhead).                                                              cuous to sufficientlyattract the reader'sattention and
                                                                         prompt him to consult the critical apparatus. The
(6) Ahbreviution.s c'onÍractions. inscriptionsin
                    und                 In                               classicalscholar Rheinhold Merkelbach has suggested
Arabic script abbreviatedwritings sometimesoccur.                        the marking of editorial substitutions by a subscript
especially the mention of dates.I have not yet come
           in                                                            asterisk instead.lT The choice of the asterisk he very
acrosscontractionsbut perhapstheseoccur too. Readers                     appropriatelyjustifles by the fact that this sign. which
cannot be supposed to understand these. and the                          is often used for reference. has the advantage of
editor will do well to fill them out. In order to indicate               drawing the reader'sattention in a very natural man-
that the portions filled out are editorial additions,they                ner to the critical apparatus.However. in conformity
are placed betweenbrackets,this time parentheses      (see               with the device proposed above for the marking of
sampleNo.     4).14                                                      lettersof doubtful reading,viz. the superscript   bar, the
                                                                         asteriskought in our inscriptions   perhapsto be placed
(7') Scrihal errors. Scribes may skip letters, or write                  not underneath the letter concerned but overhead.
superfluousor erroneousletters. Some of theseerrors                      Unfortunately, the incorporation of sub- and super-
are not made inadvertently but are due to insufficient                   script asterisksmay presentdifficulties to the typesetter.
familiarity with grammar or orthography and/or to                        I therefore suggestthat the asterisk should indeed be
contamination of whatever kind. Such errors may be                       used, but at one end of the word, and more precisely
indicative of the scribe'seducationand possiblyof the                    at the beginningof the word because the readerwill
nature of the influences which he has been exposed.
                         to                                              notice sooner.While it is true that this is an infringe-
Also. an apparent error may represent a spelling                         ment on the principle that a critical sign should be in
accepted during a certain historical period and be                       the immediate vicinity of (i.e. over. under or next to)
indicative of the developmentof the languageand/or                       the letter or text portion concerned, the deviation
the orthography. Such writings should thereforein the                    from this principle here would seemto be justifiable in
edition not be corrected but be retained as they are                     view of the limitations of typography and to causeno
(and come back to in the interpretative part of the                      inconvenience the reader (seesampleNo. 7).18
                                  F.TH, DIJKEMA. ON EDITING EPIGRAPHY IN ARABIC SCRIPT                                                         I -)-)

As to any letterspossiblyleft incompleteby the scribe,         (d) To print the visiblestrokes a partly legiblebut yet
Dow. op. cit., p. I l, has pointed out that theseshould        unidentifiable letter and to put shadingwhereverother
be considerednot as doubtful readings but as errors            portionsof that letter must be assumed have existed.
due to inadvertenceand should, editorially speaking,           This methodmay presentgreat dimculties editionsof
be treated accordingly, i.e. the incomplete letter             texts in Arabic script. which is much more varied and
should be replacedby the correct letter, an asteriskbe         complicated                                   I
                                                                            than the Greekand Latin scripts. suggestto
                                                               apply this convention possible, otherwise leavean
                                                                                      if        but            to
added and the reading on the stone be indicatedin the
                                                               emptyspace   and print a bar overhead.
critical apparatus.
   In very poorly executed  inscriptionstranspositionof
                                                               With generalreference thesevarious devicesit must
words may occurle. If the transposition is evidently
                                                               be    remarked that wherever text is somehow absent
due not to unfamiliarity with syntax but to inadver-
                                                               and no conjecturepossible,classicalepigraphistsand
tence,the editor may put the words in the right order,
                                                               papyrologistswill, in addition to the various brackets
place an asteriskat each end of the transpositionand
                                                               already usedto indicate such absences,                         place a dot for
present in the critical apparatus the reading of the
                                                               everyletter which they estimate missing,or indicate   is
original text.
                                                               the number of the missing letters by means of a
(8) Alternative readings. Previous editors or others           number: if they feel they can somehow make no such
may have suggested    alternative readings that can be         estimatethey will place dashesinstead. Owing to the
expectedto be of interest to the critical reader. The          particular nature of the Arabic script. which comprises
best place for the editor to indicate such aiternative         charactersof very different widths. which especially                             in
readings is the critical apparatus. I suggestthat the          i n s c r i p t i o n ss o f t e n w r i t t e n i n t w o o r m o r e r e g i s t e r s
reader's attention be drawn to the critical apparatus          and where the text often alternatesbetweenloose and
by means.again, of an asteriskplacedbefore the word            compressed           writing, it is in most instances            quite impos-
concernedin the edited text or, if two or more words are       sible to estimatethe number of the missing or illegible
concerned.by asterisksat both ends of the passage-             letters. Hence I suggest that in such circumstances
the use in theseinstances the asteriskhelps to avoid
                          of                                   spacesimply be left betweenthe bracketscorresponding
an undesirableincreasein the number of the critical            in size with the absentor illegibleportion. Wherever
signs used, an increasewhich, here, is also unneces-           text is availablebut the editor is unableto read it. he
sary, as the presencein the critical apparatus of the          ma1'leale corresponding                    spaceopen.tool hou'ever.              in
previous editor's siglum will tell the reader that the         order ti'r indicate that the tert in that area is not
lemnru is concerned not with a substitution or a               missing but ar ailable he places overhead the bar
transpositionbut with someoneelse'sreading. (See               indicatingdoubtful reading.
sample o. 8.)
        N                                                          With general reference to restorations, Dow. op.
                                                               cit., pp. 20ff, esp. pp 30f, proposes a scheme of
Dow recommends use.in editionsof Greekand Latin
inscriptions. yet further critical conventions,
            of                                which in         conventions to indicate their degree of probability.
editionsof our inscriptions
                          seemof little use.It may be          That schemewould also seemto be well suited for use
useful.however. mentionthem for the sakeof comolete-
               to                                              in the editions with which we are concernedhere. It
ness. recommends:
     He                                                        may sufficehere to indicate that he proposes:to add
(a) To print small superior v or vac vacat
                                    or     whereverthere is    to a restoration a small superior query mark if that
a blank spacein the original text. Given that the design of    restoration,though definitelyprobable,is not so cer-
the typical inscription in Arabic script is very regular       tain or probableas to admit no reasonable                            doubt; in
(mostly also symmetrical).that the text portions therein are   the caseof many such definitely probable restorations
usually enclosedin frames and that the text usually fills      in an edition, to use no query marks but to add to the
these completely. this convention is unlikely to serve a       edition a generalheading stating that the restorations
useful purpose in editions of our inscriptions.                are uncertain conjectures;and to presentrestorations
(b) To use capitals instead of lower case wherever indivi-     in the commentary alone if they are no more than
dually legiblelettersmake no sense  collectively.Unfortuna-    crude approximationssuggesting generalnature of        the
tely the Arabic script does not differentiatebetween lower
                                                               the restoration.
case and capitals, while differentiation between bold and
                                                                   Even though Dow does not say so, the samemethod
non-bold may be too inconspicuous,    and the use of another
clut'tLrs. K[fï, may not be possibletechnically.I suggest
                                                               may, I think, be followed in casesof uncertain resolu-
in such instances print such lettersin the same fount and
                   to                                          tions of abbreviationsor contractions.and of uncer-
type style as the others and to come back to the question in   tain editorial additions. (Uncertain readingsof erased
the interpretativepart of the edition.                         letters, as has been explained above, should, since
(c) To shadeareaswhere the surfaceis in such a condition       traces of theseletters are apparently still available,be
that it appears to have been inscribed but where attrition     overlined betweenthe double squarebrackets.)
has made the existence inscribedlettersdoubtful. Again,
                         of                                        This completesour survey of the main textual parti-
consideringthe design of the typical inscription in Arabic     cularities to be met with in our inscriptions and of
script (see above, under (a)) there is little chance of the    how they ought to be indicated in editions.Should the
editor being confronted with this uncertainty.                 printer have difficulty in typesetting one or more of
134                                    MANUSCRÍPTS OF THE MIDDLIT EAST 6 (1992)

the lesscommon brackets.as e.g. the anglebracketsor          numbers). Then. after a colon, comes the variant
the double squarebrackets,it may be useful to realize        reading or the form as it appears in the original
that most Arabic founts contain in addition to paren-        document, whichever applies. Variant readings are fol-
thesesand square brackets round and angle duckfoot           lowed by the siglum of the editor or edition concemed.22
quotes (guillentets)as well as superior and floriated        Items in a critical apparatus should preferably not be
parentheses. need be thesemight be used to replace
              If                                             separatedfrom one another by a full stop, a semi-
a kind of bracket unavailableto the printer, provided        colon. a dash or any other sign that might be considered
the editor duly notifies the reader.                         for this purpose,as this might suggestthat thesesigns
   As regardsthe layout of the edited text proper, the       form part of the portions of text quoted. Much better
following remarks may be in place. First, for easy           is it simply to print extra space,which has the additional
comparison with the original document. that layout           advantageof being more pleasingto the eye. In order
should as much as possible reproduce that of the             to avoid possibleconfusion with the edited text pro-
original text2o. Second, numbering of the lines and          per. the critical apparatus is normally printed from
stonesof which an inscription is composedis essential        smaller type and separatedfrom the edited text by
for reference.  Consideringthe fact that thesetexts are      some extra interlinear space.
generallyshort and their commentarieslong, number-               Before the edition comes the introduction, which
ing of each iine is preferable to numbering of each          should tell the reader the location. the physical featu-
third. fourth or fifth line. as is sometimesdone in          res. any detailsknown of the history of the stoneitself.
editionsof longer texts.Since.in epigraphyin Arabic          previouseditions.and who were the editors.
script, verseinscriptionswith three. four or even more           Since interpreting inscriptions often requires a cer-
hemistichs line are not uncommon.in the editions
             per                                             tain expertrse.          and since the editor himself is most
of such inscriptionssuch lines may need to overrun. In       likely of all to       possess     precisely                     it
                                                                                                                that expertise, is the
such casesversenumbering is desirablein addition to          editor's duty to have the edition accompanied a                      by
the line numbering2l. As regards the form of these           translation and a commentary.For the reader'sconve-
numberings. I have found satisfactory the method             nience and for reasons of conciseness                       certain brief
employed by myself in my edition of the inscriptions         comments need not necessarilybe included in the
in Edirne, viz. the numbering of the lines with the so-      commentary but may go in the translation between
        'Arabic' numerals (i.e. those used not in the                                                                            e.g. of
called                                                       squarebrackets.This is true for identifications.
Arabic script but in the modern Latin script). of verses     Koran citations and of towns, rivers etc. today known
with lower caseitalic lettersand of stones    with capital   under different names; interpretations of words or
roman letters.   the practicaladvantages this method
                                          of                 expressionson the stone that contain errors appa-
being that the use of non-Arabic numeralsand letters          r e n t l yd u e t o c o n t a m i n a t i o ne . g . :
precludes impression
           the             that the numberingis part of
the edited text proper. and setsoff visually'   that text.   (text:)
while to the typesetterthe use of these numerals and                                                "-.tí.r++
                                                             (translation:) ... the successorof liani.shm-i)G.Jl-
lettersin theseeditions need not presentdifficulties.                .
                                                             s h e n J . .: 2 3
   For the correct reading and understandingof verse
texts the metre may prove essential.Editors of such          and conversionsof dates to the Christian era. The
texts have in the past often contentedthemselves      with   complete yet concisecommentary discusses        any ques-
mentionrng only the names of the metres. This is of          tions of authenticity and date of the document, draws
little help to the reader, as he can hardly be expected      the reader'sattention to phenomenaof specialinterest
to have memorized all the existing metres and their          (palaeographic, prosodic, linguistic, orthographic, ter-
variant forms. Hence, when preparing his edition the         minological, religious,artistic. historical etc.) and ans-
editor should not only scan the text carefully himself       wers all of his possible primary questions, referring
but for the convenience of the reader supply the             him for further details to the main sources. The
metrical scheme,preferably at the beginning of the           commentary need not enter into questionsthe answers
edited text proper.                                          to which the readercan be expectedto find on his own
    Immediately underneath the edited text comes the         in the major dictionariesof the languageconcerned.2a
critical apparatus. This is the place where the editor          Full photographs2sas well as an index listing not
indicatesvariant readingsby previouseditors which in         only names but also conventional blessingsand epi-
his judgment deserve    consideration.It is also the place   thets and giving full references the various pheno-
where he indicates the original readings of words in         mena noted in the commentariesare indispensable         in
 which he has substitutedletters.and the order in the        any serious edition of inscriptions-without photo-
 original document of words transposedin the edited                                                    for
                                                             graphs or facsimilesthe book is useless the critical
 version. In each item in the critical apparatus the         reader,without a comprehensive                     for
                                                                                                 index useless the
 edited form, including any asterisks,comes Íirst, for       serious epigraphist.
 reference(since epigraphical texts are usually short           The arrangement of the edited texts of individual
 there will normally be no need for references lineto        inscriptionswithin a corpusis a matter needingconsi-
                                 F , T H , D I J K E M A . O N E D I T I N G E P I G R A P H YI N A R A B I C ]S C R I P T                                               135

deration. For instance, the epigraphist may prefer a                           city of our inscriptions as historical sources and
merely chronological arrangementwhile the historian                            thanks to the elegance
                                                                                                    both of the appearance and the
may prefer an arrangementby location or by histori-                            content of many of them the editing of theseis often
cal personage.  The editor must decidehere for himself                         also fascinatingand most rewarding.
which category of scholars he thinks will use his
corpus most and how their interestsare served best.
The arrangementby location is unique in having the                                                                           S,q.À,{pLes
additional advantage of leaving the way open for
endless  natural expansion,provided at least the extant                        (In the sample-textsa series of dots indicates text
epigraphy of the individual locations is presented                             availablebut here omitted.)
   Finally. it may be useful here to say a word on the
                                                                                                                             >No. l<
problemswhich this suggested     form of editorial presen-
tation may presentto the typesetters   and on how these
                                                                               Text lost, and restoredby conjecture.F.Th. Dijkema.
may be solved. When speaking, above. about over-
                                                                               The Ottoman Historical Monumental Inscriptions in
lining as an indication of doubtful reading, we have
                                                                               Edirne (Leiden 19'77),inscription No. I (the restora-
referred to the occasionalneed which this may entail
                                                                               tion is basedon contextualand historical evidenceand
of printing two successive    letters not in the form of
                                                                               on comparison with other inscriptions):
their conventionalligature but next to one another, as
if there existedno such ligature. In modern typesetting                         ... 6-:-y*cJU \ p;1p\*llo-u                            I   -\,-t-!... iU"IJlr"i
                                                                                                                                                                    ';t '.
this procedure may be possible only with certain                                                                                                : " r . ir , l
                                                                                                                                                tu u L\J               I 4!-J
technicalmeasures.    While until recentlytext in Arabic
characters was typeset on purely mechanical key-
boards having a separate key for each individual                                                                             >No. 2<
variant form of a letter (initial.medial.final. isolated).
modern electronictypesettingmachineshave one key                               Text lost and supplied from exrsting record. Ibid..
per letter only. it being the machine's  computerwhich                         inscription No. 32. line 4 (the portion supplied is
selects the proper variant form or ligature to be used                         borrowed from sel'eralchroniclesand a tedhkire):
in a certain place. This now presentsdifficuitiesin the
                                                                                                                                    -'-r* .rdt ,-' j,j-)               rlz
type-setting, the machine may refusein our caseto
set anything other than the conventional ligature.
Also. the machine may refuseto set on the right-hand                                                                         >No. 3<
side of a bracket any other variant forms of a letter
than the isolated or the final ones (to the exclusionof                        Doubtful reading. Erased text. K. Moaz & S. Ory,
the initial and medial ones) and on the left-hand side                         Inscriptions arabes de Damas. Les stèles funéraires.
any other forms than the isolated or the initial ones                          volume i: Cimetièred'al-Bab al-Sagtr (Damascus 1977),
(to the exclusionof the medial and final ones).In such                         inscriptionNo. 46:
circumstancesthe only way out is deceiving the                                      Second     t(xt

machine by inserting, in the place where the machine                                  ff]-!llr         -[tt        Lj Ár Jt -Él ll-11 *Jr 3'-ro r-t^ t
will refuse to do what is needed.the device available                                                 rrrv ,,rrr
                                                                                                         ;:::ï,.":'                        J.il ..'-:o                t1 2
to the typesetter for elongating Arabic words: the                                                    \rïY tu i                                                          :
                                                                                        it-1                                                JJ-;.Jr
elongation stroke. The insertion of this stroke will
stop the machine from selectingligatures and other                                     .r:Ér 4rj ^- Á' Gr .,. p                            er;g)l                 fy- * a
variant forms of its own choice and permit the setting                                                      ft        c Ál gl-l Jl''; Á1 +'-;                    'oj[c[il 5
of the connectedform wanted by the editor. It is true                                                                  .ó -rJt .;o-L*i' olrLi                          +J," 6
that this stroke (not necessarily  longer than one milli-                                                                                                 "!,"y
metre) mars the edition but this effect is very slight
                                                                                (Translation of the First Text:)
and is far outweighedby the advantageof the appear-
ance.in the edition. of the variant letter form actually                        I This is the tomb of the one who needsGod ímav He
wanted.                                                                         be exalted)t[he?             ]
   The picture here presentedof editing inscriptionsin                          2 l                                                                                             l
Arabic script may convey the impression that the                                3 al-SayÍï
                                                                                         [                            ]. He died
editing of such inscriptionsis an extremelycomplex                              4 on Monday the 17thof Rabi' al-Àkharof the year
and back-breakingjob. It is true that at times the                              eiehthundred
editing of inscriptionsis diÍicult. It should be remem-                         5 eighty-four [8 June 1479]. May God (may He be
bered,however,that very much of what has been said                              exalted) have mercy upon him and may God have
above concerns epigraphical particularities that are                            mercy upon whoever invokes His
rare. Instead I would sav that in view of the authenti-                         6 mercy and upon the Muslim dead. Praiseto God!
136                                                                M A N U S C R I P TO F T H E M I D D L E E A S T6 ( I 9 9 2 )

(Translation of the SecondText following the reading                                                   I w i s h t o t h a n k D r . V . L . M . M é n a g e .e m e r i t u sp r o f e s s o r
probably intended by the secondscribe:)                                                             of the School of Oriental and African Studies.London. for
                                                                                                    very kindly reading the text and helping me to presentit in
I This is the tomb of the one who needsGod (may He
                                                                                                    correct English style. I am grateful to Mr. P.F.A. Koree
be exalted)the Sheikh                                                                               and Mr. A.W. Copier. both of the former printing houseof
2        bn Ránr. He died                                                                           E.J. Brill. for advice on typographicalmatters.
3 i n t h e y e a r 1 3 3 2[ 3 0 N o v e m b e rl 9 l 3 - l 8 N o v e m b e r
t9t4l                                                                                                  1 If 'to edit critically' is meant add 'critical' before
5-6                       May God (may He be exalted) have                                             2 A n o n y m o u s( P a r i s ) .
m e r c y. . . e t c .                                                                                 3 Published in Chroniqued'Égvpte vii.Nos. l3-14
                                                                                                  ( 1 9 3 1 - 2 p.p . 2 8 5 - 7 .
(Commentary:)                                                                                          a J. Bidez et A.B. Drachmann, Emploí des signest'riti-
l. 6, Read al-amv'at. The erroneous reading on the ques etc.. Édition nout,elleby A. Delatte and A. Severyns
stonewould seemto be due not to inadvertence                                        but to ( B r u s s e l s n d P a r i s ) .
                                                                                                       s Published as volume 2 rn the series 'Greek. Roman
insufficientknowledgeof classical                            Arabic.
                                                                                                  a n d B y z a n t i n eS c h o l a r l yA i d s ' b y ' D u k e U n i v e r s i t yi n D u r -
                                                                                                  h a m . N o r t h C a r o l i n a . b u t a v a i l a b l ef r o m t h e C i r c u l a t i o n
                                         >No. 4<                                                  Manager ol Greek, Rotnan und Bl':antine Studiesat Cam-
                                                                                                  b r i d g e .M a s s a c h u s e t t T.h e b r o c h u r ei n c l u d e s m o s t u s e f u l
Resolutionof abbreviatedwriting. Dijkema. op. t.it..                                             hihliogruphíeruisonneeon the subject. Some of Professor
                                                                                                  Dou's verv preciseand appropriate definitions have been
inscriptionNo. 63. line 2:
                                                                                                 gratefulll"used in this paper.
                                    \ \ . 1 L - ( J ; ) t ( J rj ; ) [ ) ? v j . . .                   ó A. Grohmann. From the v'orld o./Arubic pctpyrí(Cairo
                                                                                                  1 9 5 2 ) ,p p . 9 8 - 1 0 9 .G e r m a n t r a n s l a t i o ni n i d e m . A r a b i s c h e
                                                                                                  Chronologie.          Arabísche        Papyruskunde          (Leiden 1966)(Hand-
                                         >No. 5<
                                                                                                 buch der Orientalistik, Erste Abteilung, Ergánzungsband
                                                                                                 I I . E r s t e rH a l b b a n d ) ,p p . 1 0 0 - 1 0 9 N . B . : i n l i n e l 9 o f p . 1 0 5
Editorial addition. Ibid., tnscriptionNo. 31, line 3:                                            read double instead of single square brackets).The section
                                                        . gí)+1 jl*: t+. .'. contains many recommendations                                             worth taking to heart not
                                                                                                 only by editors of Arabic papyri but also by editors ol
                                                                                                 inscriptionsin Arabic script.
                                        >No. 6<                                                        ' We shall deal here u,ith
                                                                                                                                                the questionswhich the reader
                                                                                                 mai, ask of an edition of an alailableinscription. ot with                          n
Editorial deletion. Ihid.. inscription No. 107. line 2 t h e s p e c i f i c u e s t i o n s h i c h h e m a v a s k o f a n e d i t i o n b a s e d
(the parentheses                                                                                                    o
                              here indicate an instanceof haplo- e r c l u s i v e l y n a s u r r , i v i n g r a n s c r i p t i o nW i t h t h e e n o r m o u s
                                                                                                                                             t                      .
graphy (seenote l4)):                                                                            m a s s e s f o u r i n s c r i p t i o n sh a t a r e s t i l l a w a i t i n gp u b l i c a t i o n
                                                                                                              o                               t
                                                                                                 in critical editions there is little chance of our Íinding the
(Texr:)                                             r
                                             u , [c Ë ( r ) . r ! : { . } : \ f i . . .          time soon for editing and publishing inscriptions whose
                          .,.                              el-flájj     'Ali ...                 texts are now only available in transcriptions.However. it
                                                                                                 may be useful to indicate here that Dow (op. cir.. pp. l2ÍJ
                                                                                                 has given some thought to the editing of this particular
                                        >No. 7<                                                  kind of text.
                                                                                                      8 Règ/es pour ëditions eÍ tradut'tions de tertes arahes
Editorial substitution.                Ibid.. inscriptionNo. 47. line 1:                         ( P a r i s1 9 5 3 )N 3 6 . 3 7 . 4 0 , 4 1 .T h e t w o s e r i e s e f e r r e d o a r e
                                                                                                                        .                                                   r             t
                                                                                                 the                          arabe' of the Association Guillaume Budé
                                                            jl o*;-,, ? ;-*l-\ :pr,
                                                                                                 and the 'Documents relatifs à I'histoire des Croisades(Tex-
(Critical apparatusi) o,,r>,:,                  : o-Li                                           tes orientaux)' of the Académie des Inscriptions et Belles
                                                                                                 Lettres. Blachereand Sauvaget's                       instruction to place kora-
                                                                                                 nic citations within floriated parentheses                           would in epigra-
                                        >No. 8<                                                  phic editions causeconfusion with the many other kinds of
                                                                                                 brackets used as critical signs. Moreover, in epigraphic
Variant reading.Ibid.,inscription No. 15, line 3:                                                editions such brackets seem to serve no useful purpose. as
                                                                                                 both the translations and the commentaries offer ample
                                                      . . . " L ; . ,L Á l , í ; l - r , . . . .
                                                                      -                          opportunity to identify the citations. The instruction of the
(Critical apparatus:)aí,1>,,: .fr:i naai, Per., Gókb., two French scholarsto vocalize fully verses,proverbs and
Ayv., On.                                                                                        citations from the Koran and from hadrth (5 39) seemsto
                                                                                                 me to serveno useful purpose either.
                                           Norrs                                                          I n i n s c r i p t i o nN o . l 5 o f t h e p r e s e n tw r i t e r ' s T h e O t t o -
                                                                                                 man hístorical monumental ínscription.sin Edirne (Leiden
     * This article is a revisedversion of a paper read at the
                                                                                                 1977)the reading of a full stop after the fifth word of the
Symposium on Textual Tradition and the Editing of Per- fourth line is indeed crucial to the understandingof that
s i a n a n d T u r k i s h T e x t s h e l d i n L e i d e n .l 6 - 1 8 O c t o b e r 1 9 8 6 . text. It is externalevidencethat has shown the need for the
                                                         F . T H . D I J K E M A . O N E D I T I N G E P i G R A P H YI N A R A B I C S C R I P T                                                r31

reading of a full stop there. It may be useful to take this                                             serveits purpose satisfactorily:it will in any event prevent
opportunity to explain that the presentpaper is basedboth                                               the appearance the edition being marred. The editor may
on experience            gained in the preparation of that edition and                                  well be advisedto discuss                this matter with the printer or the
on further research specifically into editorial technique.                                              publisher.
                                                                                                             1e For a textual error that may to some extent be
Some of the notions expoundedin this paper have therefore
been developedafter the book was completed.The book is                                                  considereda transpositionseeDijkema. op. c'it..inscription
thereforenot in all respects preciserepresentation the
                                                  a                                      of             N o . 1 2 0 .l i n e l .
                                                                                                             2 0 B l a c h è r ea n d S a u v a g e Í . p . r i r . . $ 4 7 . p r e s c r i b et h e
techniqueproposed in this article.
     10 'Lettre à M. Barbier de Meynard sur le projet d'un                                              placing. between the hemistichs of a verse. of the little
Corpus Inscriptionum Arabicarum'. in Journal Asiatíque,                                                 flower customary in Arabic manuscriptsand booki. Consi-
H u i t i è m eS e r i e . o m e x x ( 1 8 9 2 ) , p . 3 0 5 - 1 7 .
                             T                         p                                                dering the numerous signs already used in epigraphic edi-
     1t Martin L. West. Tertual criticism ancl edítorial tech-                                          tions I suggest simply leaving additional space instead.
nique upplicableto Greek und Latin le,tts (Stuttgart 1973)'                                             which. also. in the West is generally found to be more
The work is acknowledgedas authoritative also by editors                                                pleasingo the eye.
                                                                                                             21 Inscriptionsike No. 79 in D4kema. op. t'ít..require
of texts in languagesother than Greek and Latin. See e.g.
Jerome J. McGann. A c'ritiqueof modern rextttal uiticisnr                                               in this respect        much inventiveness the part of the editor.
( C h i c a g oa n d L o n d o n 1 9 8 3 ) . p . 1 f .
                                                   p                                                         2 2 T h i s i s t h e g e n e r a l l l a c c e p t e dc o n v e n t i o n i n t e x t
     1 2 S e ea b o v e .n o t e 9 .                                                                    e d i t i o n s . h e r e a s o nr l h r t h e . s i g l u t irs p l a c e dn o t b e f o r eb u t
                                                                                                                        T                                              r
     1 3 B . A . v a n G r o n i n g e n . ' P r o j e t d ' u n i f i c a t i o nd e s s 1 ' s t è -   after the variant reading is probabll' because                              editors wish to
m e s d e s i g n e sc r i t i q u e s ' .i n : C h r o n i c l u e ' E g r p t e v i i N o s .
                                                                      d                                  be consistent        uith other kinds of refèrences.                      such as referen-
 l 3 - 1 , 1 l 9 3 l - 2 ) . p p . 2 6 2 - 9 . s p .p . 2 6 4 .
           (                                    e                                                        cesto footnotes.which are conventionall.v                              piacednot prece-
     1a Haplography(writing once what should be written                                                  ding but        following the statementthey belong to.
 t w i c e ) .w h i c h i n c l a s s i c ae p i g r a p h yi 's a n e r r o r . i n O t t o m a n
                                              l                                                               23 Dijkema, op. cit.. inscription No. 78. verse c (the
epigraphy' is an accepted palaeographicrcalligraphic                                     licence         bracesthere printed are to be ignored).
 (seeDr.1kema. cit., index, s.v.).Unless an haplographyis
                         op.                                                                                  2a For the editor himself the inclusion of a translation
 apparentlydue to a mistake of the scribeit may thereforein                                              and a commentary has the advantage of allowing him
 Ottoman epigraphy,and perhapsalso in other epigraphyin                                                  another check on the correctness his edition. ln my work
 Arabic script. be considerednot as an error but as a form                                               on my own edition this has helped me to detect quite a
 o f u b b r e ii a t e d r . r i t i n g .                                                                                                                i
                                                                                                         n u m b e r o f e r r o n e o u s e a d i n g sn t i m e l
      1 s A n g l e b r a c k e t s f s o m ef o u n t s m a r t h e a p p e a r a n c e f
                                        o                                                      o              2 s W i t h r e s p e c tt o p h o t o g r a p h so f i n s c r i p t i o n sI t a k e
 t h e e d i t i o n b y t h e a m o u n t o f s p a c et h e l t a k e . T h e e d i t o r              this opportunin to conve\ for thc useol future epigraphists
 m a 1 ' b ev u e l a d r i s e dt o a s k t h e p r i n t e r t o u s ea n g l eb r a c k e t s
                       l                                                                                                                                                                   t
                                                                                                         t h e m a i n l e s s o n s l e a r n e dr i h i l e p h o t o g r a p h i n g h e e x t a n t
 v u h o s e n g l e i s a b o u t 1 3 5 " .a n d c e r t a i n l l n o t s m a l l e rt h a n
              a                                                                                          O t t o m a n m o n u n t e n t a li n s c r i p t i o n si n E d i r n e : ( 1 ) I n s c r i p -
 90..                                                                                                    t i o n s o n p h o t o g r a p h s e n d t o g r o w l e s sl e g i b l ev e r y q u i c k l y
      1 ó S e ee . g . D r 3 k e m a o p . c i t . . i n s c r i p t i o nN o . 1 0 7 .l i n e 2
                                            .                                                            a s t h e a n g l e o f t h e a r i s o f t h e l e n s o f t h e c a m e r aw i t h t h e
 a n d t h e c o m m e n t a r yo n t h a t l i n e .                                                    surface of the stone grows less than 90". Legibility is best
      r- R. Merkelbach. 'Der Stern als kritischesZeichen',in:                                            ensured if the axis of the lens coincidespreciselywith an
 Zeit.st'hri./'t      ./'iir Pupt'rologie und Epigraphik xii (1973),                                     imaginary perpendiculardropped on the verl- centre of the
 pp.2l lf.                                                                                               stone (or the portion of the stone) to be studied.(2) Detail
      18 Typesetters'founts normally include severalkinds of                                             of very worn inscriptions is best made visible with skim-
 asterisks.including large, heavy ones and smaller. light                                                ming light, that of other inscriptionswith indirect lighting
 ones. Even a rather inconspicuousasterisk will probably

To top