The Old Syriac Gospel of the Distinct Evangelists: A Key-Word-in-Context Concordance, by Jerome A. Lund, in collaboration with George A. Kiraz. [3 vols; Gorgias Press, Piscataway, 2004; ISBN 1-59333-069-3; 1-59333-070-7; 1-59333-0071-5;] xxiv, 1-952 pp; 953-1691 pp; 1692-2449 pp; hardcover.

[1] Since the 1970s there has been a steady growth in the number of biblical concordances available to readers of Syriac, beginning with Strothmann’s concordance of Qoheleth (1973), Winter’s concordance of Ben Sira and Sprenger’s of Psalms (both 1976), and then Strothmann’s word list of the Old Testament deuterocanonical texts (1988) and his monumental 14-volume concordance of the complete Peshitta Old Testament (1984, 1986, 1995), based on the text of the 1852 Urmia Bible and, remarkably, Walton’s London Polyglot of 1653-1657. Strothmann’s concordance is now in the process of being complemented by that of the Leiden Peshitta Institute, edited by Borbone and Jenner, which is based on their critical edition of the Peshitta Old Testament, and the first volume of which, covering the Pentateuch, appeared in 1997. For the New Testament, scholars were for a long time reliant on the word lists of Schaaf (1709) and The Way International (1985), before these were eclipsed by Kiraz’ Concordance to the Syriac New Testament (1993). The latest addition to these ranks of essential research tools is Jerry Lund’s impressive concordance of the Curetonian and Sinaitic Old Syriac Gospel manuscripts, which at just under 2500 pages in length is a work of truly stakhanovite proportions!

[2] The concordance is divided into three volumes, the first containing words beginning ܐ - ܗ, the second those beginning ܘ - ܠ, and the third ܡ - ܬ, plus two separate concordances of personal names and geographic names. The volumes are produced on American standard sized paper (8½" x 11", or 21.59 x 27.94cm), but with the exception of title pages and the introductory materials, the volumes are printed in ‘landscape’ format down the whole length of a page opening. That is, the columns begin at the left-hand edge of the left page and continue down to the right-hand edge of the right page. This allows sufficient space for the key word, with a significant space on either side, to be anchored at the same point in every line (as though in a separate column) with approximately five to seven words preceding and following. This of course makes it very easy to locate the word that is being sought, and provides outstanding contextualisation that is not bettered in any other Syriac concordance. To the right of the Syriac text (which is printed throughout in unvocalized estrangela type) is a reference to the manuscript and Gospel verse being cited, and then to its right a column for the insertion of each distinct Syriac form as it starts to be listed, and then finally an index number for each entry within the listing of a particular word. (All prefixed and suffixed forms of a noun are numbered together, as are all forms of a particular verb.) At the beginning of each new Syriac lemma a simple English gloss is provided which seeks to reflect the prime meaning found within the Old Syriac Gospels themselves, plus a very simple grammatical classification (N for noun, V for verb, p for preposition, etc). Some nominal and adjectival forms are provided with further grammatical classification as they are listed, but the only additional information given for verbs is their identification as peal, pael etc, which seems a little sparse compared to concordances such as Kiraz, and is particularly unfortunate given that there are some unusual verbal forms in the Old Syriac Gospels which non-specialists may find hard to identify. Unlike the Leiden concordance, however, in which different forms are all jumbled together, Lund’s work very carefully distinguishes the various morphological forms and orders them alphabetically, and this is a great help to all users.

[3] Whereas Kiraz and Strothmann list all words under their Syriac roots, Lund has explicitly followed the practice of the earlier Key-Word-in-Context concordances of Aramaic texts (Targum Neofiti, by Kaufman and Sokoloff, 1993; the Aramaic documents from Egypt, by Porten and Lund, 2002), and so lists the verbs by root (with the third radical yodh verbs listed as ܗܘܝ and ܫܕܝ etc., and not as ܗܘܐ and ܫܕܐ), but gives an alphabetical listing of all other words, with nouns presented in their emphatic / determined forms, and adjectives and those passive participles used adjectivally or substantivally, in their absolute forms. The justification given for this methodology is that by conforming to the practice of the above cited Aramaic reference works “it will allow the student of comparative Aramaic dialects to compare Syriac with those other dialects more readily”. Since the very existence of this concordance is a consequence of Jerry Lund’s work for the Comprehensive Aramaic Lexicon Project, and the support of the Project directors, and given that the Leiden concordance took the same unfortunate approach, it seems rather churlish to grumble about this arrangement, but nevertheless I do think that since the majority of users will be students of Syriac and the Syriac biblical versions, it might have been helpful to follow the lead of Kiraz and Strothmann and the major Syriac reference lexica and list all forms under the appropriate root (perhaps with an appended alphabetical key, as in Kiraz, to aid those who need assistance to track down a particular word). With the present arrangement, without the aid even of lists of root derivatives such as those provided by Jessie Payne Smith in her dictionary, users will have to work hard if they wish to identify all usages of a particular root within these manuscripts.

[4] One of the definite strengths of Lund’s concordance is that it includes listings of all uses of conjunctions, prepositions, pronouns, and particles in their alphabetic position, whereas some of these (especially ܘ, ܕ, ܒ, ܠ, etc.) are omitted from many of the other published concordances, or are relegated to appendices. The separate listing of personal and geographic names is also useful, though it might have been helpful to add English glosses to these in addition to the less familiar transcribed forms.

[5] Now, it should also be noticed that Lund was faced with a particularly daunting problem when he decided to produce this extremely welcome concordance to the Old Syriac Gospels. Whereas the Strothmann and Leiden concordances of the Peshitta Old Testament and Kiraz’ concordance of the Peshitta New Testament could work with critical editions of biblical texts based on multiple manuscripts, in which all scribal errors and idiosyncracies had been carefully removed to produce what might be termed ‘idealized’ texts, Lund had no such luxury available to him. Instead he had to work with two early, fragmentary, and highly idiosyncratic manuscripts, presenting significantly different texts, which were full of unusual orthographic forms and significant numbers of simple scribal errors. On top of this, of course, the Sinaitic manuscript is a palimpsest which is barely legible in substantial sections, despite the nineteenth-century application of chemical reagents and the late twentieth-century use of photographic imaging. Thus the editio princeps of 1894, Agnes Smith Lewis’ re-transcription of 1896 and her edition of 1910, and the 1904 edition of F.C. Burkitt, all too frequently report variant readings for the same passages. Part of Lund’s solution to these problems was to base his concordance on Kiraz’ Comparative Edition of the Syriac Gospels (= CESG; 2nd edition 2002), which had taken its own text of the Curetonian manuscript from Burkitt’s 1904 edition and of the Sinaitic manuscript from Lewis’ 1910 edition. (Although in so doing it had removed all of the small corner brackets used in these volumes to indicate that the transcribers considered a particular reading uncertain.) His second solution was to make 73 emendations to the text of the Gospel manuscripts, of which only 9 were corrections to the Curetonian manuscript. Words thus emended were included in the body of the concordance at the point determined by the alphabetic form of the emendation, although the changed form is enclosed by < >. The rejected form is also listed at this point (only) in the concordance, and is enclosed by { }.

[6] Lund provides a complete list of the emendations adopted in his concordance on pp. xi - xviii, and subdivides them on pragmatic grounds, such as ‘errors arising from metathesis of consonants’, ‘errors arising from graphically similar consonants’, ‘error by omission of a letter’, etc. He argues that “whether a modern transcriber or an ancient scribe introduced the mistake is immaterial; a textual error is still a textual error”, and so he does not seek in most cases to distinguish these sources of potential error. For anyone interested in the textual transmission of the Old Syriac Gospels there is some fascinating material here which repays closer study. There are, however, also some rather marked flaws in the methodology used to identify errors to be emended, and so I thought it might be useful to restructure Lund’s list (though keeping his numbering of items), and to provide some comments, as a first contribution to the ‘scholarly consideration’ which he invites at its beginning.

[7] A. Typographic errors in CESG alone

In the following cases a comparison with the earlier editions reveals that a simple typing error has slipped into CESG:

27.	Sin Matt 10:4	For ܣܟܪܝܛܐ	Read ܣܟܪܝܘܛܐ.
5.	Sin Luke 4:1	For ܘܕܒܬܪܗ	Read ܘܕܒܪܬܗ.
19.	Sin Luke 17:6	For ܣܡܟܐ	Read ܗܡܟܐ.
8.	Cur John 4:39	For ܕܡܢܬܐ	Read ܡܕܝܢܬܐ.
58.	Cur John 6:45	For ܟܘܠ ܡܢܗ	Read ܟܘܠ ܡܢܘ.
9.	Cur John 14:11	For ܡܝܗܡܢܝܢ	Read ܡܗܝܡܢܝܢ.

Notes:

19. For several days I thought that there was no entry in the concordance for ܗܡܟܐ, as I was simply unable to find it. By chance I then discovered it listed after ܡܢ as ܡܢ ܗܡܟܐ, and along with it forms such as ܡܢ ܐܝܡܟܐ, ܡܢ ܒܣܬܪ, and ܡܢ ܝܬܝܪ. This is far from intuitive to me, and I don’t think it should be adopted in any future concordances. (Likewise, a long search for ܐܘܪܥܐ eventually found it listed as ܠܐܘܪܥ.)

58. Lund has incorrectly emended the text here, on the basis of Sinaiticus, to ܟܘܠ ܡܢ.

[8] B. Typographic errors in Lewis and so also in CESG

In these cases a typographic error in Lewis’ 1910 edition of Sinaiticus has been preserved uncorrected in CESG:

4.	Sin Mark 16:6	For ܕܬܚܠܢ	Read ܬܕܚܠܢ.
11.	Sin Luke 9:19	For ܒܥܡܕܢܐ	Read ܡܥܡܕܢܐ.
12.	Sin Luke 10:2	For ܒܪܗ	Read ܡܪܗ.
13.	Sin Luke 10:7	For ܓܝܕ	Read ܓܝܪ.
23.	Sin Luke 15:32	For ܘܐܚܬܟܚ	Read ܘܐܫܬܟܚ.
57.	Sin John 5:6	For ܕܢܘܓܪܝܐ	Read ܕܢܘܓܪܐ.

Notes:

57. Although in Lewis’ main text the yodh is surrounded by corner brackets, in her introduction p.xxxiii she cites it as ܕܢܘܓܪܐ (as does Burkitt ad loc.), and she does not include this reading in her Appendix I where disagreements with Burkitt’s edition are listed.

The 12 corrections listed above were clearly correctly made to the text in Lund’s concordance, although given their origins in twentieth-century printing it was not necessary to justify them by comparison with other ancient texts, nor to preserve the error in the concordance (with the possible exception of 57.). These errors should also, of course, be corrected in copies of the CESG.

[9] C. Suspect transcriptions by modern scholars

In all of the following instances the text of Sinaiticus was for a long time illegible to the transcribers, and many readings in these passages remain uncertain, and so it seems reasonable that Lund’s sound emendations should be accepted.

21.	Sin Matt 24:6	For ܕܡܗܘܐ	Read ܠܡܗܘܐ.
14.	Sin Luke 11:24	For ܒܗܝܢ	Read ܒܗܘܢ.
39.	Sin Luke 23:53	For ܒܚܨܐ	Read ܒܚܝܨܐ.
15.	Sin John 6:63	For ܗܘ	Read ܗܝ.
16.	Sin John 15:4	For ܗܘ	Read ܗܝ.

Notes:

21. This proposed emendation is supported by its occurrence in Aphrahat Dem. XXI.23 (989.6).

[10] D. Scribal errors

Whilst it is possible that one or two of the following are orthographic variants, it seems more likely that they are all simple errors by the scribe of Sinaiticus.

26.	Sin Matt 8:34	For ܠܐܘܪܥ	Read ܠܐܘܪܥܗ.
60.	Sin Matt 12:36	For ܒܛܠܬܐ	Read ܒܛܠܐ.
28.	Sin Matt 12:41	For ܒܟܪܘܙܬܗ	Read ܒܟܪܘܙܘܬܗ.
1.	Sin Matt 22:19	For ܘܐܫܛܘ	Read ܐܘܫܛܘ.
29.	Sin Matt 22:30	For ܒܝܬ ܡܝܬܐ	Read ܒܚܝܬ ܡܝܬܐ.
61.	Sin Matt 26:36	For ܥܡܡܗܘܢ	Read ܥܡܗܘܢ.
30.	Sin Matt 26:69	For ܓܠܝܠܐ	Read ܓܠܝܠܝܐ.
72.	Sin Mark 6:40	For ܘܚܡܫܝܢ	Read ܘܚܡܫܝܢ ܚܡܫܝܢ.
2.	Sin Mark 9:18	For ܘܡܪܚܩ	Read ܘܡܚܪܩ.
3.	Sin Mark 14:11	For ܦܥܠܐ	Read ܦܠܥܐ.
31.	Sin Mark 14:70	For ܕܓܠܝܠܐ	Read ܕܓܠܝܠܝܐ.
34.	Sin Luke 5:4	For ܡܠܠܗ	Read ܡܡܠܠܗ.
6.	Sin Luke 9:1	For ܥܠܝܐ	Read ܚܝܠܐ.
33.	Sin Luke 17:37	For ܢܬܟܫܘܢ	Read ܢܬܟܢܫܘܢ.
56.	Sin Luke 18:22	For ܙܒܢܢ	Read ܙܒܢ.
7.	Sin Luke 19:44	For ܕܣܥܘܪܢܟܝ	Read ܕܣܘܥܪܢܟܝ.
44.	Sin Luke 22:48	For ܠܒܪ ܕܐܢܫܐ	Read ܠܒܪܗ ܕܐܢܫܐ.
63.	Sin John 6:52	For ܚܕ ܥܡ ܚܕܐ	Read ܚܕ ܥܡ ܚܕ.
71.	Sin John 20.1	For ܕܡܓܕܠܝܐ	Read ܕܡܥܓܠܐ.

Notes:

29. This occurs in a section that could not be read before Lewis’ 1910 edition, and so it is possible that this is a transcriptional error rather than a scribal error.

30. This same emendation is also listed by Lund as 51.

72. In this instance an entire word is being added that does not exist in the manuscript, on the grounds that normal grammatical practice requires it. I am sure this is right, but it should be noted that this is a significant step away from simply recording the extant text.

With these readings Lund’s policy of listing the word under the emended form but also presenting alongside it the original erroneous form is highly appropriate.

[11] E. Non-standard orthography

There are good grounds for thinking that all of the following are orthographic variants, and not scribal errors. Some are attested in other early Syriac texts, others follow well known patterns for such variants. It is a mistake to attempt to conform these words to later standards, or to those taught in the church schools and monasteries of the Edessa region. (Indeed, the extent of the unusual orthography found in these manuscripts is clear from the most cursory browse through the pages of the concordance, and it is a mine of information that has as yet been little explored by linguists.) In this table Lund’s proposed emendations are listed in brackets in a column placed before the original readings which are to be kept.

24.	Sin Matt 3:4	(For ܘܡܐܟܘܠܬܗ)	Keep ܘܡܐܟܠܬܗ.
25.	Cur Matt 3:14	(For ܠܘܬܝ)	Keep ܠܘܬ.
65.	Cur Matt 13:16	(For ܘܛܘܒܝܗܝܢ)	Keep ܘܛܘܒ.
17.	Sin Mark 6:9	(For ܟܘܬܝܢܝܢ)	Keep ܟܘܬܘܢܝܢ.
52.	Sin Mark 10.32	(For ܒܐܘܪܚܐ)	Keep ܒܐܪܚܐ.
35.	Sin Luke 12:5	(For ܐܝܢ)	Keep ܐܢ.
37.	Sin Luke 13:3	(For ܬܬܘܒܘܢ)	Keep ܬܬܒܘܢ.
38.	Sin Luke 14:10	(For ܬܫܒܘܚܬܐ)	Keep ܬܫܒܚܬܐ.
42.	Sin Luke 14:12	(For ܦܘܪܥܢܐ)	Keep ܦܪܥܢܐ.
43.	Sin Luke 16:6	(For ܘܟܬܘܒ)	Keep ܘܟܬܒ.
40.	Sin Luke 19:17	(For ܐܝܘ)	Keep ܐܘ.
53.	Sin Luke 19:36	(For ܒܐܘܪܚܐ)	Keep ܒܐܪܚܐ.
45.	Sin Luke 23:16	(For ܐܪܕܝܘܗܝ)	Keep ܐܪܕܘܗܝ
46.	Sin Luke 23:16	(For ܘܐܫܒܩܝܘܗܝ)	Keep ܘܐܫܒܩܘܗܝ.
47.	Sin Luke 23:22	(For ܐܪܕܝܘܗܝ)	Keep ܐܪܕܝܗܝ.
48.	Sin Luke 23:22	(For ܘܐܫܒܩܝܘܗܝ)	Keep ܘܐܫܒܩܘܗܝ.
41.	Sin Luke 24:26	(For ܠܬܫܒܘܚܬܗ)	Keep ܠܬܫܒܚܬܗ.
50.	Sin John 7:51	(For ܠܘܩܕܡ)	Keep ܠܩܕܡ.
59.	Sin John 10:9	(For ܢܚܐ)	Keep ܢܝܚܐ.
54.	Sin John 14:5	(For ܐܘܪܚܐ)	Keep ܐܪܚܐ.
49.	Sin John 14:18	(For ܐܫܒܘܩܟܘܢ)	Keep ܐܫܒܩܟܘܢ.

Notes:

24. As Burkitt noted in his helpful notes on the grammar and syntax of these manuscripts (cf. Burkitt II.40), Sinaiticus has a marked tendency to omit the letter waw in prepositions, nouns, and verbs, (as do other early Syriac texts, cf. Drijvers and Healey, The Old Syriac Inscriptions of Edessa and Osrhoene [1999], p.23), and this particular form is also attested in Assemani’s Acta Martyrum (1748) II.74.

25. Silent final waw and yodh are occasionally, if rarely, omitted from the ends of words in the Old Syriac Gospels and elsewhere. Cf. Taylor, The Syriac Versions of the De Spiritu Sancto by Basil of Caesarea (1999), Index Orthographicus, p.183-195.

65. Strictly speaking this is not a matter of orthography, but rather the word in the Curetonianus appears to be the common abbreviation which is used in numerous manuscripts for the full form suggested by Lund. (These abbreviations often occur at the ends of lines, but I have not been able to check this here.)

52., 53., 54. Another example of a word that can lose its waw in early orthography, and which is also attested in the Peshitta of Matt 20:30 in Pusey and Gwilliam’s cod.36, and is cited by Payne Smith from the Roman edition of Ephrem.

35., 40. Both of these words are found in manuscripts and in the lexica with their defective forms.

38., 41. These are not plural forms but an orthographic variant of the singular that is also found in a sixth century manuscript of Basil on the Holy Spirit (cf. Taylor, loc. cit.) and Isa 52:1 and 61:3 in 5ph1 (cf. VTS III.1 p.xvi), Isa 42:10 in 6h3, and in 7pk18 (op. cit. p.xviii).

42. This is yet another example of defective spelling, which happens to be attested in a second-century inscription, As37:7 (Drijvers and Healey, op. cit., p.108)

45. - 48. Burkitt II.54-56 lists numerous occurrences of verbs in Sinaiticus with object suffixes that do not conform to later orthographic standards (far more than are marked for emendation by Lund), and in Taylor, op. cit., further examples of these are listed from various sources.

50. By way of comparison, it should be noted that the scribe of Sinaiticus has a liking for forms such as ܠܩܒܠ and ܠܩܒܠܗ.

[12] F. Possible textual variants / unjustified emendations

The following are rather a rag-bag of readings. Some of them are early Syriac textual readings which may well pre-date the writing of the two surviving Old Syriac manuscripts, and others are capable of being read as coherent variants within their context, or are actually supported by known variants in other early New Testament witnesses. In all cases they seem to me to be instances where it is not appropriate for a concordance to emend the text (although a note that they may well be errors, or need emending by critics, is of course perfectly reasonable in an introduction). The practical consequence of such emendation is that a curious reading such as 68., ܘܫܛ is now only to be found by reading the introduction and knowing that it must be looked for under ܘܫܕܪ! If it is an error it is an interesting one, and should be left in its appropriate context for other scholars to consider.

64.	Cur Matt 1:21	(For ܠܥܡܗ)	Keep ܠܥܠܡܐ.
18.	Sin Matt 4:10	(For ܠܒܣܬܪܝ)	Keep ܠܒܣܬܪܟ.
20.	Cur Matt 11:6	(For ܕܠܐ)	Keep ܐܠܐ.
66.	Cur Matt 14:13	(For ܒܠܚܘܕܘܗܝ)	Keep ܒܠܚܘܕܝܗܘܢ.
22.	Sin Matt 25:40	(For ܙܥܘܪܐ)	Keep ܙܥܘܪܝ.
36.	Sin Mark 6:26	(For ܕܢܗܦܘܟ)	Keep ܕܢܗܦܟ.
73.	Sin Mark 8:34	(For ܘܢܐܬܐ ܒܬܪܝ)	Keep ܘܢܐܬܐ.
32.	Sin Mark 10:4	(For ܕܢܟܬܘܒ)	Keep ܕܢܟܬܒ.
67.	Sin Mark 10:49	(For ܕܢܩܪܘܢܝܗܝ)	Keep ܕܢܩܪܒܘܢܗ.
55.	Sin Mark 11:1	(For ܘܠܒܝܬ)	Keep ܠܒܝܬ.
62.	Sin Mark 13:8	(For ܕܚܒܠܐ)	Keep ܕܚܒܠܬܐ.
10.	Sin Luke 1:5	(For ܐܒܝܐ)	Keep ܐܒܝܡ.
68.	Sin Luke 1:53	(For ܘܫܕܪ)	Keep ܘܫܛ.
69.	Cur Luke 24:20	(For ܩܫܝܫܐ)	Keep ܩܫܝܫܝܢ.
70.	Sin John 13:5	(For ܘܡܫܘܐ)	Keep ܘܡܫܦܪ.

Notes:

64. The Greek, and Sinaiticus and Peshitta, read ‘his people’, and Lund would so emend, arguing that this is ‘an internal Syriac development, produced by scribal lapse’. Burkitt II.257, on the contrary, states that ‘the variation in the suffix shews that the change was not due to a simple graphical error’, and he points to Luke 2:10 where Sinaiticus and Peshitta have ‘world’ for Greek ‘people’. (Lund lists 9 further internal Syriac textual developments on p.xix.) If sense, literal or theological, can be made of the reading, as here, it should not be emended, because otherwise this begins to result in the conforming of the Old Syriac Gospel text to some hypothetical Greek norm.

20., 66., 22. Possibly an error, but possibly not as an intelligible rendering can be given in context, and thus to be left without emendation.

73. In this instance an error does seem likely, but again it is not certain, and does not warrant emendation in the body of a concordance.

32. This could be an orthographic variant, or the use of an Aphel form to indicate the writing of a decree or formal document, and in either case no emendation is necessary.

67. In the parallel passage in Luke 18:40 Jesus commands that the blind man be brought to him, and not just be ‘called / summoned’, and it is noteworthy that in the Persian Diatessaron (III.33) it appears to be the Lukan wording of this verse that is inserted into the Matthaean version of the episode.

62. Although this is the only attested feminine form of this word in Syriac I think it is significant that as acute a lexicographer as Brockelmann felt that it should be included in his Lexicon, 210b. (It might also be noted that the feminine form does exist in various Jewish Aramaic dialects, such as JBA, where it has the meaning ‘destruction’.) This may be an error, or it may be a precious attestation of a rare word, but it should be left in the text of a concordance for readers to decide.

10. Not only does the Hebrew text of the Old Testament occasionally read ‘Abiyam’ instead of ‘Abiya’, but the Old Latin manuscripts e l also read abiam in this very verse. It is far from clear, therefore, that this is a Syriac scribal error.

68. The reading ‘he despised the rich as being empty/worthless’ is certainly remarkable, and it may be an error. Alternatively it could be an internal Syriac development of encratite / ascetic motivation, as witnessed in numerous other passages, and so it would be a mistake to conform it to the Greek text.

69. Lund appears to take the Curetonian reading here as a plural absolute form ‘elders’, and so for grammatical reasons emends it to the plural emphatic / determined form. He adds ‘This seems better than interpreting the form ܩܫܝܫܝܢ as “our elders”’. The Greek here, however, literally reads ‘the chief priests and our rulers’, and so it seems highly likely that this word should indeed be read as ‘our elders’. (Another instance of the Curetonian having been brought into closer conformity with the Greek text than Sinaiticus or, occasionally as here, the Peshitta.)

70. This proposed emendation is also quite radical. Lund argues that ‘it is clearly a textual error, probably by a modern transcriber who was unaware of this meaning [‘wipe’] of ܫܘܐ in Syriac’. But in this passage the transcribers seem to have been able to read the text without too much difficulty (although this is relative), and they re-read it on several subsequent occasions. Also, given that the Peshitta of this passage also reads ܘܡܫܘܐ, the transcription of an alternative reading was not a consequence of ignorance but a conscious recording of what they believed they could see. It is possible to make some sense of ܘܡܫܦܪ in this context, and so the reading should stand. (If an alternative reading were required for other purposes I think a more plausible reconstruction of the original text would be ܘܡܟܦܪ, using the same root employed in the Harklean.)

[13] I have devoted far more space than is normally possible in a book review to a consideration of Jerry Lund’s list of emendations because I found the issues they raised very interesting, and I hoped that this new analysis would stimulate further discussion. By my very crude ‘back of an envelope’ estimate, however, this affects 73 out of some 170,000 lines of text, and even in these 73 lines Jerry has explained very clearly what he has done, and has preserved both the original reading and his emendation. It would be quite wrong, therefore, to interpret this discussion as a critique of the larger work. On the contrary, I firmly believe that the publication of Jerry Lund’s concordance marks the beginning of a whole new era of text-critical and linguistic study of the Old Syriac Gospels, and I for one would like to thank him enthusiastically for the invaluable resource that he has placed within our hands.