THE ALLEGED "CORRUPTION"
OF THE HEBREW TEXT.

This Is Appendix 93 From The Companion Bible.

PAGE   1
    In modern commentaries we very frequently meet with the objectionable word "corruption" used of the Hebrew text of the Old Testament.

   As specimens of this feature of modernism, the following are taken at random from one of the latest commentaries:—
  1. This "probably signifies not only a new paragraph but a later hand."
  2. This "leads to the conclusion that there is some original corruption of the Hebrew text."
  3. "The text in this verse is extremely difficult to interpret; and no satisfactory translation can be given of it."
  4. "The Hebrew of this verse seems to be so corrupt that there is no satisfactory meaning to be obtained from it."
  5. "It is certain that the original text must be corrupt."
  6. "It is better to regard it as being in some way a corrupted text . . . but is now unintelligible."
  7. "These three verses are extremely corrupt, and it is probably impossible to restore the text with any certainty."
  Such remarks abound; and very few pages are free from them. There is a continual running confession of inability to understand the Hebrew text. Like the schoolboy who always thinks "the book is wrong", modern critics never seem to suspect that the difficulty lies with themselves and not with "the Book". We must accept their confession, whatever the explanation may be.
  The object of this Appendix is to show that those who are so ready to speak about "corruption" can have little or no knowledge of the Massorah, or of its object.
  We have explained its character somewhat in Appendix 30. We now propose to point out that its one great special aim and end was to make such "corruption" impossible.
  Well knowing the frailties and infirmities of human nature, those who had charge of the Sacred Text hedged it round on all sides with regulations and information called the Massorah, because it was meant to be "a fence to the Scripture", and because it should be, thus, next to impossible for a scribe to make a mistake in copying it.

  Some general facts are given in Appendix 30 (which should here be consulted); but further particular features are now added from Dr. C. D. Ginsburg's four large folio volumes, which contain the Massorah so far as he has been able to collect, arrange, and transcribe the writing in smaller characters at the top and bottom of every page of most of the accessible manuscripts containing it.
  I.  All the letters of the Hebrew text were counted: not as a piece of mere curiosity, but that the number of each letter in each book being thus known to the scribe he might easily check his work, and ascertain whether one letter had escaped or got over "the fence". He was informed how many Alephs (Hebrew aleph= A) there should be, how many Beths (Hebrew beth= B), etc., in each book respectively.

  II.  There are five consonants, which when they occur at the beginning of a word must have a dot within them, called a Dagesh. This dot in no way affects the meaning of the word.
  In certain positions, other than at the beginning of a word, these five letters may, or may not, require this Dagesh. Now, each of these dots was safeguarded; for one might so easily be omitted or misplaced: hence, the scribe was assisted by an instruction that, in cases where any of these five letters should not have a Dagesh, he must make a small mark over it, called a Raphe. This again in no way affected either the sound or the sense; but it reminded the scribe that in these cases he had to do one thing or the other. He must write it (if the letter were, say, a Beth (Hebrew beth= B) either Hebrew bethdagesh or Hebrew bethraphe.

  III.  Again: certain letters have come down with the text, from the most ancient times, having a small ornament or flourish on the top: for example, we find
          Aleph (= A) with 7 Taagin Hebrew letter Aleph with 7 Taagin. 966 bytes          
          Beth (= B) with 3 Taagin Hebrew letter Beth with 3 Taagin. 926 bytes          
          Gimel (= G) with 4 Taagin Hebrew letter Gimel with 4 Taagin. 933 bytes          
          Daleth (= D) with 3 Taagin Hebrew letter Daleth with 3 Taagin. 914 bytes          
  These ornamented letters were quite exceptional, and implied no added meaning of any kind: but, so jealously was the sacred text safeguarded, that the scribe was informed how many of each of the letters had these little ornaments: that is to say, how many Alephs (Hebrew aleph= A), and how many Beths (Hebrew beth= B), etc., had one, two, three, or more.

  These ornaments were called Ta'agim (or Tagin), meaning little crowns. The Greek-speaking Jews called them little horns (Hebrew keranoth) because they looked like "horns". The Authorized Version and Revised Version rendering of keraia (Greek = horn) is "tittle", which is the diminutive of "title" and denotes a small mark forming such title.
TOP                           PAGE  1                         PAGE  2                         BOTTOM                           APPENDIX   LIST                           HOME

THE ALLEGED "CORRUPTION", ETC. (cont.).

PAGE   2
  Modern commentators, and even the most recent Dictionaries of the Bible, still cling to the traditional explanation that this "tittle" is the small projection or corner by which the letter Beth (Hebrew beth= B) differs from kaph  (Hebrew kaph= K); or Daleth (Hebrew daleth= D) differs from Resh (Hebrew resh= R), etc.
  But the Massorah informs us that this is not the case, and thus, tradition is quite wrong. We give a few examples showing how even these little ornaments were safeguarded:—
  Rubric Hebrew aleph, § 2 (Ginsburg's Massorah, volume ii, pages 680-701) says: "Aleph with one Tag: there are two instances in the Pentateuch (Exodus 13:
5, Hebrew aleph in 'asher (= which), and verse 15Hebrew aleph in 'adam (= man).
  Rubric Hebrew aleph, § 3, says: "There are seven Aleths (Hebrew aleph= A) in the Pentateuch which respectively have seven Taagin".
  Rubric Hebrew beth, § 2, notes Beth (Hebrew beth= B) with one Tag, as occurring only once (Exodus 13:
11, yebi'aka = brings thee).
  Rubric Hebrew beth, § 3, notes Beth (Hebrew beth= B), as occurring in four instances with two Taagin: namely, Genesis 27:
29 (ya'abduka = may serve thee); Genesis 28:16 (bammakom = place); Exodus 7:14 (kabed = is hardened); Exodus 23:23 (vehayebusi = and the Jebusites).
  Rubric Hebrew beth, § 4, gives four instances  where  Beth  (Hebrew beth= B) has three Taagin: and so on, through all the alphabet, noting and enumerating each letter that has any Tagin: thus safeguarding the sacred text, so that not one of these little ornaments might be lost.
  It was to these Taagin the Lord referred in Matthew 5:
18, and Luke 16:17; when He said that not only the smallest letter (Hebrew yod= Yod = Y), but that not even the merest mark or ornament (Tag) should pass away from the Law until all things should come to pass. So that our Lord Himself recognized these Taagin, which must have been in His Bible from which He quoted.

  IV.  In cases of spelling, where a word occurs a certain number of times, but in one or two cases with a slightly different spelling (where, for example, one was with a short vowel and another with a long or full vowel), these are noted, numbered, and thus safeguarded.
  The scribe is not left to imagine that some of these are incorrect, and so be tempted to correct the smaller number by making them conform with the larger number of cases in which the word is spelt differently. It is needless to give examples of such instances.

  V.  Where a certain word or expression occurs more or less frequently in varying forms, these are all noted, numbered, and distinguished. For example, the word bayith (= house); its occurrences with different vowels and accents are all safeguarded.
____________________________
  ¹ Ginsburg gives verse 13; but volume ii shows that it is verse 15.
  So with its occurrences with certain prefixes and suffixes: for example "in the house", six occurrences, where the letter Beth  has  a  Sheva  (Hebrew bethsheva) are safeguarded against thirty-two where it has a Pathach (Hebrew bethpathach) instead.
  So with its combinations with other words: two are noted as being "in this house which is called" (Hebrew beth, § 244); nineteen as being "into the house" (Hebrew beth, § 245); twice "and within the house" (Hebrew beth, § 246); four times "and the house of", and "and into the house of" (Hebrew beth, § 247); twice "the house of her husband" (Hebrew beth, § 249); "house of Elohim" five times without the Article: these five exceptional cases being thus safeguarded against the forty-eight occurrences where Elohim has the Article (Hebrew beth, § 251).
  In nine instances "House of Elohim" is followed by the demonstrative pronoun "this": but, in five cases this pronoun is the Chaldee language dek (Ezra 5:
17; 6:7, 7, 8, 12), and in four cases it is edenah. These latter are thus safeguarded.
  The occurrences of the expression "the house of Israel" are noted separately in the  Pentateuch  and  the  Prophets  (Hebrew beth, §§ 254, 255); and in Hebrew beth, § 256, these are further distinguished from the expression "the sons of Israel" (the words beyth, "house of", and b
eney, "sons of", being much alike in Hebrew).
  "Shearing house" is noted as occurring twice (Hebrew beth, § 258), and "house of restraint" as occurring three times (Hebrew beth, § 257).
  "Jehovah Adonai" is noted as occurring 291 times; but the fewer occurrences of "Adonai Jehovah" are safeguarded against the more usaul form (Hebrew yod, § 178).
  Jehovah our Adonay is safeguarded against the more usual form "Jehovah our Elohim" (Hebrew yod, § 179).
  In the same way, the following exceptional phrases are distinguished: "Jehovah the Elohim", "Jehovah Elohim of", "Jehovah Elohim Z
eba'oth", "Jehovah Elohim of heaven", "Jehovah my Elohim", etc., etc.
  The expression "the sins of Jeroboam", which occurs fifteen times, is in ten instances followed by "the son of Nebat". The shorter phrase is thus exceptional; and the scribe is warned not to make any of the five like the other ten by adding "the son of Nebat".

  These examples might be enumerated by hundreds from Dr. Ginsburg's Massorah; but enough are here given to show how the Massorah was indeed "a fence to the Scriptures".

  In the face of these facts one might smile (if the case were not so serious) at the readiness of modern critics to use the word "corruption" whenever they have to admit that they cannot understand the text as it stands. We have no reason to doubt the truth of their confessions; but it is better, and easier, and happier, and safer to believe God.
TOP                                                           PAGE  1                                                                     PAGE  2

Appendix List

CLICK HERE TO GO HOME
therain.org 1999
website security