Skip to main content
Internet Archive's 25th Anniversary Logo

Full text of "Number of Letters in the Pentateuch"

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 



Alexander Marx 
Jewish Theological Seminary of America 

The number of letters in each of the books of the Pentateuch 
is not found in any edition of a Masoretic Bible before Gins- 
burg's. The latter derived his figures from his MS. 6, a compi- 
lation of the early part of the nineteenth century by one Simon 
Silberberg who intended to publish and collected subscribers for 
his work in 1828-1834. 1 

Professor Blau 2 pointed out that there are very curious mis- 
takes in these numbers; to Leviticus exactly the same number 
of letters is given as to Genesis, and to Exodus 20,000 less than 
to any other book. He further proved their inaccuracy by com- 
paring these figures with the sum of those given to the weekly 
portions in the Masora. Ginsburg nevertheless repeats his 
incorrect statements in his new edition of 1908. 3 Yet there is 
Masoretic information available on this point which has been 
overlooked by both scholars. Richard Simon in his Histoire 
Critique, book I, chapter XXVI, gives the data from a Bible 
MS. written in Perpignan 1300. The MS. he refers to is 
undoubtedly the one described as no. 7 of the Catalogue des MSB. 
Hebreux et Samaritains de la Bibliotheque Imperiale, Paris, 
1866, formerly MS. Oratoire 5 and thus coming from the Library 
of the institution with which Simon was connected. For the 

1 Ginsburg, Introduction to the Massoretico-Critical Edition of the 
Hebrew Bible, London, 1897, p. 762-5. lb., p. 110-11, a specimen of Silber- 
berg 's method is given which shows that he followed to some extent the 
scheme given by Hartmann, Linguistische Einleitung in die Biicher des 
A. T., Bremen, 1818, p. 133, which Hartmann thought to be that of the Old 

W.Q.S., XVI, 359-60. 

3 In the same way Ginsburg in the IV vol. of his Masora takes no cog- 
nizance of Blau's proof (J.Q.B., VIII, p. 352 seq.) that Saadiah's poem 
on the letters contains only the number of times they occur in Prophets and 
Hagiographa excluding the Torah. 


sake of convenience I quote his words from the English transla- 
tion : 4 

"I have nevertheless seen a Manuscript of Perpignan wherein 
was this part of the Massoret with several others. And that no 
one may doubt hereof I will set them down as I read them 
reckon 'd up in this manuscript Copy. There are 12 Parscioths 
or great Sections in Genesis. There are 43 of those which are 
call'd Sedarim or Orders. There are 1534 Verses, 20,713 
Words, 78,100 Letters, and the midst of this Book consists in 
these words, Ve al harveka tihieh, in Chap. 27, Ver. 40. There 
are five Points, (these are Points made on the top of some letters 
mention 'd by S. Jerom.) 5 Exodus has 11 Parscioths, 33 Seda- 
rims, 1209 Verses, 63,467 Letters, and these words, Elohim Lo 
Tekallel, in Chap. 22. Ver. 27. are in the very middle of this 
Book. There are in Leviticus 10 Parscioths, 25 Sedarims, 859 
Verses, 11,902 Words, 44,989 Letters, and these .Words, Vehan- 
nogea bibesar, in Chap. 15. Ver. 7. are the middle words. There 
are in Numbers 10 Parscioths, 33 Sedarims, 1288 Verses, 16,707 
Words, 62,529 Letters, and these words, Ve haia-is asher ebehar, 
in Chap. 17. Ver. 5. are the middle words. There are in Deu- 
teronomy 10 Parscioths, 31 Sedarims, 9055 Verses, 16,394 Words, 
54,892 Letters, and the middle words of this Book are ve Ascita 
Alpi Hadavar, in Chap. 17. Ver. 10. " e 

My attention was drawn to this quotation by a reference in an 
old, otherwise useless grammatical work by Georg Sharpe 7 in 
which the number of letters is discussed in order to prove the 

4 A Critical History of the Old Testament. Written originally in French 
by Father Simon, Priest of the Congregation of the Oratory. And since 
translated into English, By a Person, of Quality [E. Hampden]. London, 
1682, p. 362. He refers to the same MS. in chapter XXIII, p. 147, with 
the words: "I have seen a MS. which had 24 Books of the Bible which had 
been writ at Perpignan in the year 1300 in a neat Character. ' ' 

5 On the margin he adds the reference : Hieron. Quest. Hebr. in Gen. 

c The Samaritans also have counted words and letters of their Pentateuch ; 
see Heoraica, IX, 222-3 ; JBL., XXV, 40-2. 

'Two Dissertations: I. Upon the Origin, Construction, Division, and 
Kelation of Languages. II. Upon the Original Powers of Letters; wherein 
is proved from the Analogy of Alphabets, and the Proportion of Letters, 
that the Hebrew ought to be read without Points. To which is added, The 
Second Edition, enlarged, of a Hebrew Grammar and Lexicon, without 
Points. By Gregory Sharpe, LL.D. London, 1751. 


unreliability of the Masoretes. It is interesting to compare his 
method with that of Professor Blau 8 who repeated the same 
investigation using as one of the three texts on which he bases 
his calculations the very edition consulted by Sharpe. As 
Sharpe's book is not very common it may not be superfluous to 
reprint his words (p. 69-72) in full : 

"The Masorets are said to have counted every letter in every 
book, and to have left us the exact number of times that each 
letter occurs throughout the Bible. It must have been very diffi- 
cult for them, not having the artificial arithmetic of the moderns, 
to manage very large numbers; and, to be sure, they did not 
imagine that any man would be so idle as to reckon them up 
after them, to prove the truth or f alshood of their calculations : 
And if the proportion of some of their letters had been more 
accurately expressed, the reader would not have been troubled 
with the following supputation : 

"In the bible published by Desmarestz at Amsterdam, Anno 
1701, without points, each leaf contains four columns, each 
column 51 lines, and each line, at a medium, 21 letters; which 
is less than the truth. To determine the number and proportion 
of letters, I cast up four columns, in four different books. Part 
of the 27th chap, of Genesis, making a complete column, contains 
1110 letters ; part of the 23d chap, of Jeremiah 1090 ; the book 
of Obadiah 1121. For a medium, if you add these numbers 
together, and divide the total by four, you have 1092, which is 
more than 21 letters to a line. After this I went through the 
bible, page by page, and allowed 21 letters for each line that I 
supplied, in order to make the number of lines in every column 
full 51, where the beginning of books, of chapters, or of sections, 
occasioned any break or defect in the lines: Allowing also for 
six blank columns after the Pentateuch, for four blank columns 
between Kings and Isaiah, for six before the Psalms, and two 
at the end : And as some few pages of names and numbers did 
not contain 21 letters in each line, I reckoned them at 19 letters 
the line. On the whole, in this way of supputation, I deducted 
20 leaves, 1 column, and 18 letters from the 293 leaves there are 
in the whole book: And then multiplying the remainder by 4 
for the number of columns, and the product by 51 for the num- 

8 J.Q.S. VIII, p. 345-46. 


ber of lines, and that again by 21 for the number of letters, the 
total came out 1,168,083 ; which, although it be less than the 
truth, is 352,803, more than 815,280, the number of the Masorets. 
And if you were to take away 400 columns, or 100 leaves from 
the 293 there are in the book, the remainder alone will exceed the 
Masoretic number by 11,532. Hence, to use the words of Bux- 
torf, Luculenter perspicitur, quanta horum hominum fuerit 
industria, quam laboriosum studium, quantusque zelus, ut integ- 
ritatem vel in minimo APICE inviolatam conservarent. 

' ' Pere Simon says, he saw a manuscript at Perpignan in Spain 
(sic), in which there was an account of the number of letters 
contained in the Pentateuch, very different from that of Rabbi 
Saadia, which is called the Masoretic number. In the Penta- 
teuch only, according to that MS. there are 303,977 letters. 
And if, according to my manner of computing them, you allow 
13 columns and the two blank columns after the Pentateuch, 
which belong to p. 75 and four lines, which is near the truth, and 
deduct this from the product of 75 X 4 X 51 X 21, you will have 
305,151 = the number of letters in the Pentateuch. This is 
1,174 more than in the MS. of Perpignan ; but then it is as near 
as possible : For if I had allowed 50 instead of 51 9 for the number 
of lines, and 20 for 21, the number of letters in a line, the total 
would have come out 19,057 less than in the MS. If I had 
allowed 51 lines in a column, and only 20 letters to a line, the 
number would have been 13,357 less than in the MS. If I had 
allowed 21 letters to a line, and but 50 lines to a column, the 
number woud have been less than that of the MS. by 4,811. The 
difference therefore of 1,174 is not to be regarded. It is rather 
a proof that neither of the sums are very far from the truth ; but 
which is nearest will not be readily determin'd by those, who 
know how difficult it is to sum up such a number of letters. ' ' 

Sharpe comes back to this question in the preface to his Lexi- 
con p. 9-11 and here he gives some really valuable information 
as far as these minute questions are of interest. 

"I have lately received an account of the numbers taken by 
Meyer Cohen, a Jewish teacher, learned in his own language, 
under the direction and for the satisfaction of a gentleman 

9 Blau deducts four lines less and his result (il). p. 34(5) is accordingly 
slightly different (3052351. 


skilled in the mathematic sciences, who required the sum of the 
letters to be taken or cast up in small parcels, verse by verse ; 
which was done, and the entire sums are as follow : 








28052 . . . 







...( 1634) 



...a. . 

" ' I 25095 






. . .(21612) 


....18125 .. 





w . . 



( 9873). 

. 9854... 

... J.. 

' ' !• 14111 

( 4352). 

. 4257... 

... |.. 

( 8616). 

. 8610... 


' ' I 11960 

( 3362). 

. 3350... 






. ... 7186... 

...( 7187) 

i. . 

. ... 7032... 

...( 7034) 

( 3975). 
( 831). 

. 3976... 

. 834... 


' " I 4810 


. ... 4694... 

...( 4701) 

( 2929). 

. 2925... 


' ' I 3992 

( 1033). 

. 1067... 


? . 


...( 2200) 

} . 

. ... 2109... 

...( 2105) 

D. . 

. ... 1833... 

...( 1843) 

a. . 

. ... 1802... 

...( 1812) 

Total 304805 

The number of Letters in 

Genesis 78064 

Exodus 63529 

Leviticus 44790 

Numbers 63530 

Deuteronomy 54892 

Total 304805 

According to my account. .305151 

Difference only 346 

According to the MS. of 

Perpignan 303977 

Difference 828 

10 For the sake of comparison I add in brackets the numbers of Silberberg 
as put together from Ginsburg's Massorah by Blau (ib., p. 352-3). 


Evidently Meyer Cohen did the same work which Silberberg 
repeated seventy years later and with very similar results. Con- 
siderable differences we only find in a few instances : with "? 
they amount to 40 ; with 1 to nearly 100 ; and with !1 to 900 ; 
in a few other instances differences amount to about 20. The 
only very large differences we find in reference to D. Here 
Ginsburg gives the number 1634 while the number found by 
Cohen amounts to 26,344. Blau had pointed out the evident 
incorrectness of the former figure in this instance. But his simi- 
lar claims in reference to JJ f D ~ 1 are not substantiated by the cal- 
culations of Cohen.