Some remarks preliminary to a biblical chronology
There are a number of points where modern Biblical chronologies disagree with each other. For those who do not assume gaps in the genealogies of Genesis 5 and 11 the largest area of dispute is the assessment of the varying merits of the Masoretic Text, Septuagint (LXX) and Samaritan Pentateuch. Evidence is given which suggests that the Septuagint and Samaritan traditions have suffered chronological revision in the course of transmission. The various alleged uses of Septuagint and Samaritan chronology by New Testament writers are surveyed and found to be explicable by ways other than by supposing that the New Testament writers followed the Septuagint or Samaritan Pentateuch. Future areas of chronological research are considered.
There have been many attempts to derive from the Bible a date for the creation of the world. Some believe that the Bible does not give data indicating a date for creation. It seems to the author that the Bible does give such data, since it gives the vital information required to calculate the length of specific periods. It does this by indicating the ages at which people begat successive generations (Genesis 5 and 11). Where there is danger of confusion1 we are given the essential information we need to create a chronology.2 This paper is not an attempt to create a Biblical chronology. Rather, it is an attempt to clarify some issues in Biblical chronology, so that some common ground can be established amongst creationist chronologists.
First, we list in approximately descending order of importance the causes of difference among modern chronologists:
Some believe that the genealogies of Genesis 5 and 11 contain gaps. However, assuming the correctness of the Masoretic figures in Genesis 11, even the addition of 100 missing generations to that list would add little more than 3,500 years of post-Flood time. This is because the members of the list beget at 35 years of age or under.
The figures in Genesis 5 and 11 are partially different in the Masoretic Text, Septuagint3 and Samaritan Pentateuch. This makes a difference of considerably under 2,000 years at the most.
It has been argued that the ages of begetting in the genealogies indicate the time when someone’s firstborn was begotten, not necessarily the time when the next person in the list was begotten. This has been argued particularly strongly on the basis of Acts 7:4 for Terah’s begetting Abraham.7
Every BC date chronology has to work back in time from a fixed point.10,11 The now popular chronology of Thiele uses astronomical data to achieve its fixed date of 763 BC from which Thiele calculates his scheme with the reign of Jeroboam beginning in 931/930 BC.12 This is later by about 45 years than the chronology of Ussher. This is because Thiele’s chronology has coregencies (where the reigns of kings overlap) of up to 24 years in one reign,13 while earlier schemes had interregna (times between kings where there was no monarch) of up to 22 years.14
Unless it is assumed that every genealogical figure’s wife gave birth to the next figure in the list on his birthday then we may have to allow for a small lack of resolution in a chronology. At the most this would be under a year for each individual.15
A similar problem to the above concerns whether ‘two years after the flood’ (Genesis 11:10) means ‘two years after the flood ended’, ‘two years after the flood began’, ‘in the second year after the flood ended’,16 or ‘in the second year after the flood began’.
There are other disagreements, but these are the main ones.
I do not intend here to resolve all the differences between chronologists. The issues are too complex for a general answer. Much has already been written about issue (1). While it is my personal conviction that there are no grounds for supposing gaps in the Genesis genealogies (other than perhaps the one of Cainan—see Luke 3:36), this belief cannot be proven in the way that the solution to disputes (2)–(6) can be demonstrated. I here turn to issue (2)—that of the use of the Septuagint and Samaritan Pentateuch in chronology.
There have been modern creationist supporters of the chronology of the Septuagint.17,18 The Septuagint chronology gives a greater age for the Earth than the Masoretic Text does, while the Samaritan Pentateuch gives a shorter period from Creation to the Flood, and a longer period from the Flood to Abraham than the Masoretic Text does. It is probable that at least some of the support for the Septuagint among creationists is derived from the desire for there to be more time in Earth’s history. Here are set out the figures of the three texts under comparison for the ages in Genesis 5 and 11 (see Table 1 below).19-22
We must ask the question as to which versions have changed the figures? This can be answered by means of both external and internal evidence. I examine here whether any of the versions has a proven history of changing chronological information.
|NAME||Age at begetting||Remaining years of life|
|LXX||Masoretic Text||Samaritan Pentateuch||LXX||Masoretic Text||Samaritan Pentateuch|
Examination of the Septuagint for evidence of alteration of the text
Most Christians nowadays know very little about the Septuagint. There is a general understanding that the New Testament has used the Septuagint, and some therefore assume that it must be in some way inspired. Most people are not aware that the Septuagint as a whole shows significant differences from the Masoretic Text from which English Bibles are translated. These are spread all over the Septuagint and would take a long time to list. A few illustrations of these differences will be given:
The Septuagint Exodus displays a different arrangement from the Hebrew of much of the section describing the Tabernacle. It also does not have a large section of chapter 38.
The Septuagint 1 Samuel does not have 17:12–31, 17:55–18:5 and 18:17–19.
The Septuagint 1 Kings has 14 additional verses after 2:35, and 11 additional verses after 2:46.
The Septuagint Job has substantial pluses and minuses from the Masoretic Text, including five extra verses after 2:9 and a five verse epilogue after 42:17. 21:28–33, 22:13–16, 26:5–11, 28:14–19, 31:1–4, 34:28–33, 36:5b–11, 36:29–37:5a and 39:13–18 are marked with an obelisk in manuscripts, indicating that they were not in copies of the Septuagint before Origen (ca. AD 185–ca. AD 254).
The Septuagint Proverbs has five extra verses after 24:22. Immediately after this is placed 30:1–14, then 24:23–34, then 30:15–31:9, then 25:1–29:27, before the book ends with 31:10–31.
The Septuagint Jeremiah lacks about 3,100 words of the Masoretic Text and is about 6/7 of the length of the Masoretic Text. The oracles against the nations (Masoretic Text chapters 46–51) are in a different internal order and are situated after 25:13, not at the end of the book.
From this very incomplete list of variations it should be obvious that merely to say that the New Testament uses the Septuagint, and therefore its figures are more reliable, is to raise more questions than it answers.
Instead we must ask ourselves what is the origin of these differences. There are two options:
These differences were introduced in translation.
These differences existed in Hebrew texts from which the Greek was translated.
In fact there is certainly some truth in both of these statements. D. W. Gooding has shown how the differences mentioned above in I Kings 2 are an alteration introduced when or after I Kings was translated into Greek.23 At the same time one of the Dead Sea Scrolls (4Q Jer b) is a Hebrew manuscript showing an omission of Jeremiah 10:6–8 and 10:10 just like the Septuagint. This is the case even though other Hebrew manuscripts of Jeremiah from the Dead Sea Scrolls do not generally support Septuagint omissions. A further question must be asked at this stage: when we have conflicting Hebrew manuscripts, which are original?
I here investigate the possibility that the variations were introduced either in the Septuagint or in the Hebrew manuscripts from which the Septuagint was translated.
Revision at the Greek Stage?
Ironically those who, like Setterfield, calculate chronologies on the basis of the Septuagint do not do this consistently. Since at least the time of John Lightfoot, there has been the habit of calling chronologies ‘septuagintal’ when in fact they are hybrid. This is because the Septuagint chronology for the whole Old Testament is different from that of the Masoretic Text. For instance, in I Kings 6:1 according to the Septuagint, Solomon’s Temple was founded in the 440th year after the Exodus, not in the 480th year as in the Masoretic Text. The chronology of the Israelite and Judean kings is so different in the books of Kings in the Septuagint and the Masoretic Text that the Septuagint preserves two accounts of Jehoshaphat’s reign, one after 16:28 and one in 22:41–51. On linguistic grounds the former is certainly the earlier Septuagint form.24 These differences can be shown to result from the fact that both the Masoretic Text and the Septuagint seek to record each king’s reign in chronological order. The alterations originate in variant understandings of the chronological data of Omri’s reign in I Kings 16:21–28. Shenkel believed that there had been systematic revision of the Masoretic Text to a new chronology.25 This view was ably refuted by Gooding, who showed that in fact the Masoretic Text was original.26,27 In fact, Gooding has shown that the Septuagint of I Kings has a tendency to ‘correct’ in a pedantic way any chronology with which the translator or reviser perceived there to be a slight problem.28 This sort of fussiness is something we will see in the next section. On the grounds of analogy between Genesis and I Kings, it is very possible that the figures in Genesis could have undergone revision by a Greek translator or reviser.
Revision at the Pre-Greek Stage?
The fact, however, that the Samaritan Pentateuch agrees with the Septuagint in some differences from the Masoretic Text makes it much more likely that chronological differences go back to a pre-Greek stage of the Septuagint, that is, to the Hebrew manuscripts from which the Septuagint was translated. This is because the Samaritan Pentateuch is in Hebrew and therefore attests to the existence of chronologically variant Hebrew manuscripts. However, we should not look upon the two possibilities of Greek and Hebrew variants as mutually exclusive. With our present knowledge there is nothing to preclude the hypothesis that figures in Genesis 5 and 11 were variant in the Hebrew stage and revised further in the Greek stage of the Septuagint tradition. But if the variants do go back to a Hebrew stage, we must ask which is original: the Masoretic Text or the Septuagint’s Hebrew predecessor?
We will see that the same fussiness about chronology as evidenced itself at the Greek stage of the Septuagint I Kings is present in what was probably the Septuagint’s Hebrew predecessor in Genesis:–
The Septuagint to Genesis 2:2 reads,
‘And God finished on the sixth day his works which he did and rested on the seventh day from all his works which he did.’
Here the Masoretic Text records the ‘seventh’ day in both parts of the verse. The variant, however, seems to be pre-Greek since the same variant shows up in the Samaritan Pentateuch and in the Syriac Peshitta version.29 However, the change can be explained as part of a pedantic tendency of some scribe. Fearing lest talk of God finishing His works on the seventh day should lead people to think that things were still being created on part of that day, he altered the figure.
Again the Septuagint to Genesis 2:19 reads,
‘And God formed again from the ground all the beasts of the field …’
The word ‘again’ is not in the Masoretic Text, though it is in the Samaritan Pentateuch. It is clearly added as part of a tendency to make explicit what is implicit. There were two possible understandings of Genesis 2:19 (another creation, or referring back to the one before), and the reviser has made explicit his view. The evidence of analogy suggests that pre-Greek revision has taken place from a text like the Masoretic Text to one which is more explicit about chronological issues.30
Chronological revision of the Samaritan Pentateuch?
As suggested by the two examples above (Genesis 2:2 and 2:19) the Samaritan Pentateuch has also been revised in some chronological issues. We must ask, however, when this happened. Although all Samaritan Pentateuch manuscripts are after the 13th century AD and it might be tempting therefore to place the revision at a late stage in transmission, the evidence favours an early date for revision. This evidence comes in two forms. Firstly, agreements between the Samaritan Pentateuch and the Septuagint cannot be explained on the basis of any late change. Secondly, Hebrew texts from the Dead Sea Scrolls (called ‘Proto-Samaritan’ texts31) seem to show variant Hebrew texts which agree with the Samaritan Pentateuch in places, but which do not show the peculiar sectarian revision of the Samaritans.32 The Samaritan Pentateuch, according to Purvis, has undergone two stages of revision. Firstly, a revision which involves explanation of the text, substitution of easier word forms for more difficult ones, and clarification of perceived ambiguity; secondly, a sectarian revision in which elements particularly justifying Samaritan as opposed to Jewish religion were introduced.33,34 The Samaritan Pentateuch is agreed by all scholars to contain revisions to make it easier to read. This revision, or more probably tendency to revise over a period of time, needs to have begun at a stage before the Septuagint was translated, and may well have continued after that stage, given the number of times that the Samaritan Pentateuch and the Septuagint do not agree in chronology.
I will give a few more examples of chronological easings in the Samaritan Pentateuch that do not correspond to the Septuagint. This shows not only that the whole process is complex, but also, by showing that the Samaritan Pentateuch is sometimes alone in its chronology suggests that revision is a tendency of this version.
In the Samaritan Pentateuch to Genesis 50:5 Joseph says,
‘My father made me swear before he died, saying …’
The words ‘before he died’ are not in the Septuagint35 or the Masoretic Text. Obviously some reviser was concerned to specify the time of the event, but shows a rather pedantic character—as if anyone would understand the passage as suggesting that Joseph had received a communication from his father after he died!
In the Samaritan Pentateuch to Genesis 50:23 we read,
‘And Joseph saw for Ephraim children of the third generation. Also the children of Machir son of Manasseh were born in the days of Joseph.’
Here the Masoretic Text and the Septuagint attest the reading ‘on the knees of Joseph’ instead of ‘in the days of Joseph’. Clearly a reviser of the Samaritan Pentateuch has substituted what he perceived to be a clear chronological marker (‘in the days of Joseph’) for one which was not so clear (‘on the knees of Joseph’).
In the Samaritan Pentateuch to Exodus 13:6 we read,
‘For six days you shall eat unleavened bread, and on the seventh day is a festival to the LORD.’
The Masoretic Text has ‘seven days’. The Samaritan Pentateuch is obviously trying to ease a problem it felt in the Masoretic Text. The change is very similar to that mentioned above in Genesis 2:2.
The clearest argument that we cannot simply select extracts from the Samaritan chronology without accepting the whole is the nature of the actual figures in Genesis 5. Whereas if the figures in the Masoretic Text of Genesis 5 are simply added together one concludes that the Flood took place in AM 1656, the Samaritan figure for the Flood from Genesis 5 is AM 1307 (Anno Mundi, ‘year of the world’). In the Masoretic Text using the same method of calculation Methuselah dies in AM 1656, perhaps during the Flood. In the Samaritan chronology, however, Jared, Methuselah and Lamech all die in exactly AM 1307, presumably all in the Flood. The Samaritan chronology demonstrates a number of things:
The Samaritans did not regard Jared, Methuselah and Lamech as spiritual heroes.
The Samaritans believed that simple addition of the numbers in the genealogies would produce a correct chronology. At the time of their revision, which is almost certainly pre-Christian, they did not believe there were gaps in the genealogies, and they did not believe that there were cumulative imprecisions in such calculations (see issue (7) in the introduction).
The figures in their chronology form a coherent whole from which one cannot simply select one figure and leave the rest.
The originality of the Masoretic Text
The evidence we have considered so far has suggested that chronological revision has been a characteristic of both the Samaritan Pentateuch and the Septuagint transmission. We must here consider the question of textual families. The Samaritan Pentateuch and the Septuagint share a number of readings that they do not share with the Masoretic Text. It is often stated that the combined witness of the Septuagint and the Samaritan Pentateuch against the Masoretic Text is strong because they are independent. However, Bruce Waltke, who studied the Samaritan Pentateuch for a doctoral dissertation,36 has argued that we should rather consider the combined witness of the Masoretic Text and either of the other two texts as strong, while the combined witness of the Samaritan Pentateuch and the Septuagint is of little textual significance.37 This is because, as a broad generalisation, the Samaritan Pentateuch and the Septuagint may be visualised as two sub-branches from one branch of the textual tree, while the Masoretic Text is the other branch. Agreement from less related witnesses is more likely to represent the original than agreement from close sub-branches, which may only represent their immediate predecessor. When this axiom is applied to the chart of the readings of the three versions above, it will be seen that there are good grounds for ascribing the originality to the Masoretic Text. The Masoretic Text frequently agrees with one, but not the other of the Samaritan Pentateuch and the Septuagint. Take for instance the ages in Genesis 5. The Masoretic Text and the Samaritan Pentateuch are in agreement for the first five names until Jared. Thereafter, except for the ages of Enoch and Noah, who were indisputably good, the figures diverge and the Samaritan Pentateuch has calculated the ages so that the remaining three antediluvians (Jared, Methuselah and Lamech) die in the Flood. At this point the Masoretic Text agrees more with the Septuagint.
There are, however, some outstanding questions. The Bible is a consistent whole and the alleged use of Septuagint or Samaritan Pentateuch readings in the New Testament needs to be examined.
The New Testament and The Septuagint chronology
Does the New Testament use the Septuagint Chronology in Galatians 3:17?
‘This I say: a covenant confirmed before by God [in Christ] the law which comes after 430 years does not invalidate so as to annul the promise.’
This text talks of a 430-year period between the confirmation of a promise and the giving of the law.
The Masoretic Text—‘Now the sojourn of the children of Israel, which they dwelt in Egypt, was 430 years.’40
The Septuagint—‘Now the sojourn of the children of Israel, which they dwelt in the land of Egypt and in the land of Canaan,41 was 430 years.’
The Samaritan Pentateuch—‘Now the sojourn of the children of Israel and their fathers, which they dwelt in the land of Canaan and in the land of Egypt, was 430 years.’
First, the variation in word order between the Septuagint and the Samaritan Pentateuch suggests that it is they, not the Masoretic Text, that have been tampering with the text.42 What is common to them both is the Masoretic Text, and therefore according to Waltke’s theorem that where the Masoretic Text agrees with one of the others then the Masoretic Text is to be preferred, we may suppose that both the Septuagint and the Samaritan Pentateuch are wrong. Secondly, an explanation of their text is easy to come up with on the grounds of the fussiness previously mentioned. The concern evidently comes from Genesis 15:13, which was understood to talk of a four hundred year period of oppression by another nation. The Septuagint and the Samaritan Pentateuch texts, not content with viewing Genesis 15:13 as an approximation (400 for 430) decide to alter the period in Exodus 12:40 to include some of the stay in Canaan too. Since the Septuagint Exodus 12:40 refers to the stay of the ‘children of Israel’ (= children of Jacob), the figure there probably does not contain any time for Abraham and Isaac since they both precede Israel (= Jacob) genealogically. Therefore the idea of a 215-year sojourn in Egypt preceded by 215 years in the land of Canaan finds no justification in the Septuagint. Though it is an ancient view that Paul used the Septuagint in Galatians 3:17, the Septuagint does not say what people assert it says. The Septuagint probably intends to speak of a thirty-year stay of Jacob’s children in Canaan, before they went down to join Joseph, and thereafter of a 400-year stay by the descendants of Jacob. The still later addition of ‘of their fathers’ in the Samaritan Pentateuch may represent the view that the period stretches from Abraham to the Exodus.
A more probable solution to the question has been suggested by Keefe in an M.Th. thesis.43 Keefe argues that the whole patriarchal period is viewed by Paul in Galatians 3:17 as the time of promise, since the promises, we are frequently told in the Old Testament, were to Abraham, Isaac and Jacob.44 In Galatians 3:17 ‘430 years’ refers to the period from the end of the time of promise to the time when the law was given. In addition it may be noted that just after the possession of the land has been repromised to Abraham in Genesis 15, in Genesis 15:16 he is first told of four ‘generations’ or four hundred years of oppression which is going to come upon his descendants. Few consider that the four ‘generations’ begin immediately at the time when Abraham was addressed in Genesis 15:16. Paul, then, does the same as Genesis 15 in considering a period after Abraham, but not commencing immediately after him. Even if this solution is not correct, the extreme textual difficulties of saying Paul simply used a Septuagint chronology must be dealt with. In addition, Galatians 3:16 talks about ‘promises’ (that is, plural) of the land to Abraham and ‘to his seed’. Galatians seems to be including the other occasions when promises were given (for example, Genesis 13:15 and 17:8) not simply considering Genesis 15. There is therefore no reason to assume that Paul’s 430-year period must begin at the time of Genesis 15 rather than at other times of promise.
The Number of Generations in the Sojourn
An additional argument which is sometimes used regarding the length of the sojourn of the children of Israel in Egypt is the number of generations involved. Some believe that Numbers 26:58–59 suggests that Moses was the great-grandson of Levi via the line: Levi-Kohath-Amram-Moses. This view is difficult but not impossible, since it entails Moses’ grandfather having produced 8,600 male descendants by the time of the Exodus (see Numbers 3:28). It seems more likely that what is described in personal terms in Numbers 26 is using the common figure of representing a descendant’s actions as those of their ancestor. Compare I Chronicles 7:22 and particularly verse 23 which talks of Ephraim and his wife, and also Judges 1:2–4. If this is the case then Numbers 26 does not suggest the shorter time limit that some people think it does.
Does the Addition of the Name ‘Cainan’ in Luke 3:36 affirm the validity of the Septuagint Chronology?
In Luke 3:36 the name ‘Cainan’ appears between that of Arphaxad and Shelah. This represents an additional name not present in the Masoretic Text of Genesis 11, but present in the Septuagint, though not the Samaritan Pentateuch. In examining the reason for this there are several separate questions:
(1) Was Cainan’s name originally in the genealogy in Genesis 11?
(2) Could it be secondary to the genealogy of Genesis 11 and yet historically true? In other words, could it be a non-canonical, but historically true addition, the information of which is canonised by Luke’s use of the name in his list? This might open up the possibility of more non-canonical but true gaps in the genealogy.
(3) If the name of Cainan is correct in the list, does that mean that the Septuagint’s details about his biography are correct?
(4) If the details are not correct, then there are a further two possibilities:–
(a) we have no figure for when he gave birth,
(b) we need to calibrate the Septuagint age given at ‘begetting’ to be 100 years less to fit in with the pattern of the Masoretic Text which gives ages as 100 years less than the Septuagint, that is, he begat Shelah at 30 not 130.
I have argued for a tendency to revise chronology in both the Septuagint and the Samaritan Pentateuch. However, the addition of a name need not be part of any such revision since it is neither pedantic, nor clarificatory. Furthermore, the omission of a name in a list of formulae which begin and end in the same way is a highly common error of transmission. If such an error has occurred in the Masoretic Text, the Septuagint may be original at least in the issue of having the extra name. The addition of a name is not primarily a chronological issue, and it may therefore be a distinct variant from the chronological data attached to that name, and unconnected to the variants that I have argued are certainly secondary. Without appealing to such a copyist’s error in the Masoretic Text (and I would prefer not to) the Cainan problem is still unsolved to my knowledge.
The New Testament and the Samaritan Chronology
Does Acts 7:4 Imply Approval of the Samaritan Chronologies?
In Stephen’s speech in Acts 7:4 he says,
‘Then going out of the land of the Chaldeans, he [Abraham] dwelt in Haran. And from there, after his father died, He [God] moved him [Abraham] to this land in which you now dwell.’
Some say that Stephen is following the chronology of the Samaritan Pentateuch.45 We will see, however, that his speech represents a careful reflection on the text of Genesis. It may, of course, be disputed whether or not his interpretation is inerrant, or whether Luke is simply inerrantly recording what Stephen historically said.
In the Samaritan Pentateuch of Genesis 11:32 Terah dies at the age of 145, whereas in the Masoretic Text (and the Septuagint) he dies at the age of 205. According to both the Masoretic Text and the Samaritan Pentateuch Abraham was 75 when he left Haran.46 Why are there differences between the Samaritan and the Hebrew chronology? The answer is simple. Genesis 11:26 (Masoretic Text and Samaritan Pentateuch) reads, ‘And Terah lived seventy years, and begat Abram, Nahor and Haran.’ The Samaritan Pentateuch believed that Genesis 11:26 indicated that Terah was 70 when Abram was born, and that therefore he would only be 145 years when Abram left Haran. Although the text of Genesis 11–12 is not explicit it may be read as implying that Terah had died when Abram left Haran (see below). The chronological figure for Abram’s life could not be altered since his ages are referred to later, for example, Genesis 17:1. The Samaritan Pentateuch therefore altered the figure for Terah’s age as the only way to avoid the perceived contradiction.
How Might One Infer from the Text that Terah Had Died When Abram Left Haran?
I have asserted above that the view of Stephen that the text implies that Terah had died before Abram left Haran might be derived from a close reading of the Genesis text. This does not necessarily mean a correct reading, but it seems indisputable that at the time of the New Testament people read texts carefully for information which was not explicit in them. I will run through the sequence of events in Genesis:
11:27 Terah begets Abram, Nahor and Haran. The verse, by giving an age for Terah’s begetting, need not imply that his children were triplets any more than Genesis 5:32 does when it gives a similar formula for Noah’s begetting Shem, Ham and Japheth. Nor is the order of names necessarily according to age.47
11:28 Haran dies while his father is still alive, in Ur of the Chaldees. Haran did not die a juvenile, since he had daughters (Milcah and Iscah verse 29) and a son (Lot verse 31).
11:29 Abram and Nahor marry. Haran’s daughter is old enough to marry her uncle. This may imply a considerable difference in age between the brothers Nahor and Haran.
11:31 After the marriages Terah, clearly in charge, takes Abram, his grandson Lot, and his daughter-in-law Sarah,48 and sets off for Canaan. They do not reach Canaan, but settle in the city of Haran.49
11:32 Terah dies. This is the only formula in the genealogy where it is said that someone died, and since it is not told in the main body of the genealogy (that is, Genesis 11:26), great emphasis is placed on the fact that Terah did not make it any further than Haran.
12:1 This verse either begins ‘And the LORD had said’, or ‘And the LORD said’. The extent to which this grammatical form suggests a pluperfect is disputed among Hebrew grammarians.50,51 The verse, however, suggests that it was said while Abram was still in the land of his birth, that is, Ur of the Chaldees. This favours the translation ‘had said’. Stephen’s chronological information is gained from this observation when he says (Acts 7:2–3), ‘The God of glory appeared to our father Abraham when he was in Mesopotamia, before he dwelt in Haran and said to him, “Get out of your land … ’
12:4 Here we read of Abram’s journey in obedience to God’s call. The narrative of 11:32 is resumed.
12:5 Uncle Abram, not grandfather Terah, is in charge of, and takes, Sarah and Lot. This indicates a change of control from Terah to Abram and thus may implicitly indicate the death of Terah. In this verse Abram takes ‘all their property which they had gained, and all the people which they had gained in Haran’. If Terah were still alive we might expect the property and people to remain with him in Haran.
We see then that all of Stephen’s chronological material might have been gained by him from a close reading of the Biblical text, and need not be derived from the Samaritan Pentateuch. The information may have been derived by following some contemporary exegetical practice in part attested by the Samaritan Pentateuch.
An extra 60 years?
If Stephen’s data are reliable, there is an obvious implication for chronology—we cannot simply use Terah’s age of 70 at Genesis 11:26, as just one more ‘begetting’ age to be used in our addition. Ussher and Lightfoot supposed that we must add 60 years into the chronology. This is because if Abraham was a mere 75 when he went from Haran to Canaan, and he left immediately after Terah had died (aged 205), then he must have been born when Terah was 130 (205 minus 75). This means that Abraham was born 60 years later than the date we would gain by simply using the age of 70 given for Terah in 11:26 when he begat his three sons. James Barr, who does not believe that early Biblical chronological data correspond to a historical reality, opposes the addition of sixty years to the system on exegetical grounds:–
‘Gen. xi. 26 says that when Terah was seventy years old … he became the father of Abram, Nahor and Haran. Now the natural meaning of this, I submit, is that these three were born in that year; no reasonable person would take it otherwise. Abraham was born when Terah was seventy … and when he was 75 he migrated to Canaan … Terah was still alive, and died later on, still in Haran, sixty years after Abraham had departed …
Since the migration into Canaan is the datum point for the following stages of the chronology, it means that from here onwards Ussher’s system runs sixty years … behind the one that the Hebrew naturally suggests …’52
Barr believes that Bible passages may be composed of sources which contradict each other, and that therefore it is not proper to make a reconstruction of the chronology of a section of the Bible by taking into account data from another part of the Bible which is from a different origin. Let us, for the sake of argument, assume Barr’s premise that New Testament data, or data from widely differing parts of the Old Testament, are not to be included in a chronological reconstruction. There are still sufficient data from the Old Testament itself to demonstrate that his comment ‘no reasonable person would take it otherwise’ is an overstatement. We have four indications from the text of Genesis that when one person begets three sons at a certain age, it is not necessarily to be supposed that all three were begotten simultaneously:
- Nahor marries his brother’s daughter (Genesis 11:29).
- Ham was the youngest of Noah’s sons (Genesis 9:24).
- Japheth is the oldest of Noah’s sons (Genesis 10:21).
- Shem was not 102/103 but 100 when he begat Arphaxad (Genesis 5:32, 7:6, 11:10).
Although it would be possible for scholars who believe in a source-critical approach to Genesis (that is, J and P) to divide these up into different sources, the similarity of the material suggests that any such division would be a much more expensive hypothesis than simply to suppose that Genesis 5:32 and Genesis 11:26 do not have to talk of the begetting of triplets. But for a distaste for synthetic (harmonistic) readings, both Genesis 5:32 and Genesis 11:10 would be regarded as from the same source, since they are both part of toledoth ‘genealogy’ material.
We are left then with the question of whether to add 60 years to our chronology. The problem here is that I cannot find any Biblical reason that demands that Abraham left Haran as soon as his father died. If one is not found then the addition of 60 years is merely a probable guess. Can we preclude on the grounds of the information in Stephen’s speech the possibility that in his understanding Terah died aged 205 and that Abraham waited for a few years before making his journey from Haran? I merely ask the question. It would seem more likely that if Abraham were following guidance to a certain land and had temporarily stopped in Haran, perhaps because of the ill-health of his father, he would not wait there long after his father died. If he had stayed in Haran much longer, then he must have been born even later than when Terah was 130 and a period of more than 60 years would be added to Biblical chronology. At any rate, we can say that the addition of 60 years is the minimum if Stephen’s chronological data are taken into account. If his data are not taken into account, construction of a coherent chronology seems easier. In considering this problem we should remember that often chronological data in the Old Testament are given in quite subtle ways, and only painful exegesis will reveal them.53
Conclusions and Recommendations
Any Biblical chronology that is proposed must be consistent in its methodology. A chronologist must not simply pick between varying texts in an ad hoc way, since the Septuagint and the Samaritan Pentateuch show evidence of systematic schematisation within themselves.
We have seen that some of the formidable challenges of Biblical chronology do permit reconciliation. While some of the comments here are tentative, it is hoped that they may provoke others to consider the material more closely. There remains work to do on exegesis of the Biblical text. The ‘Cainan’ problem has not yet been resolved.54 In particular, we need as much data as we can get about possible extra-Biblical sources for chronology. We need to compile a list of all calculations of the age of the Earth by people using Jewish or Christian sources up to the first few centuries AD.55 These may give us a clue as to what text and exegesis individuals were using at a given time. These details may, or may not, play a part in an ultimate harmonisation of all Biblical data in this area.
Pete Williams has been a graduate student at the University of Cambridge, England, where he has received an M.A. (Cantab.) in Classics and Hebrew, and an M.Phil. in Hebrew Studies, and has a Ph.D. on the Syriac (Peshitta) translation of the Old Testament. At time of writing, he was carrying out post-doctoral research on weapons in Ancient Hebrew at the Faculty of Divinity, University of Cambridge. Return to Top.
I would like to thank Dr Leslie McFall and Mr Stephen Lloyd for their helpful comments on a previous draft of this paper.
References and Endnotes
- Genesis 5:32 and 9:24 together indicate that we cannot assume that Shem was born when Noah was exactly 500 years old. Return to text.
- Genesis 11:10—‘And he [Shem] begat Arphaxad two years after the flood’. Return to text.
- That is the Greek Old Testament probably translated in the third and second centuries BC. In the case of the Pentateuch more evidence suggests a third rather than second century origin. Return to text.
- Ussher, J., Chronologia Sacra, ch. viii. Reprinted in: C.R. Elrington (ed.), 1847–1864. The Whole Works of the Most Rev. James Ussher, Hodges and Smith, Dublin, Vol. 11, pp. 580–598. Cited in: Barr, Ref. 11. Return to text.
- Hoehner, H., 1969. The duration of the Egyptian bondage. Bibliotheca Sacra, 126:306–316. Return to text.
- Davis, J.J., 1986. Moses and the Gods of Egypt, 2nd edition, Baker Book House, Grand Rapids, Michigan, pp. 156–158. Return to text.
- Whitcomb, J.C. and Morris, H.M., 1961. The Genesis Flood, Presbyterian and Reformed, Philadelphia, pp. 479–480. Return to text.
- Except for in the wild but early manuscript Codex Bezae Cantabrigensis (approximately 5th Century AD.) and possibly in the damaged manuscript P75 (3rd Century). Codex Bezae is the prime representative of the so-called ‘Western’ text of Acts. This text is approximately 8 or 9 per cent longer than other texts of the book of Acts, and contains many other substantial differences from the texts from which all English Bible translations are made. Return to text.
- Codex B (Vaticanus) omits the parallel passage in I Chronicles from ‘Arphaxad’ in 1:17 to ‘Shelah’ in 1:24. This passage is, however, in Codex A (Alexandrinus). Cainan’s name is therefore absent from B and present in A. This raises the possibility that if B had been preserved for Genesis 11 (see Ref. 19) the name ‘Cainan’ would be absent, and that B might have contained here the original Septuagint reading. The name ‘Cainan’ is also present in the Septuagint of Genesis 10:24. Return to text.
- From this angle we must recognise that we cannot build a BC date world chronology exclusively from the Bible despite Whitelaw’s assertion to the contrary: Whitelaw, R. L., 1986. Recent creation and world-wide Flood: the perfect agreement between Biblical chronology, recorded history, and other extra-biblical geochronometers. Proceedings of the First International Conference on Creationism, R. E. Walsh, C.L. Brooks and R. S. Crowell (eds), Creation Science Fellowship, Pittsburgh, Pennsylvania, Vol. 1, pp. 129–134. We can only build an AM (Anno Mundi ‘year of the world’) chronology for the Old Testament. From the Bible alone there is no way of telling the exact time gap between the Testaments. Even interpretations of Daniel 9:24ff which attempt to cover this gap require information external to the Bible to determine the length of a ‘week’. Return to text.
- Barr, J., 1984–1985. Why the world was created in 4004 BC. Archbishop Ussher and Biblical chronology. Bulletin of the John Rylands University Library, 67:575–608, especially p. 579. Return to text.
- Thiele, E.R., 1983. The Mysterious Numbers of the Hebrew Kings, 3rd edition, Academie Books, Grand Rapids, Michigan, pp. 67–78 and p. 217. Return to text.
- That of Azariah. See: Thiele, Ref. 12, p. 219. Return to text.
- Crocket, W. D., undated. A Harmony of Samuel, Kings and Chronicles, reprinted 1985, Baker Book House, Grand Rapids, Michigan, p. 295. This interregnum appears between Jeroboam II and Zechariah, kings of Israel. Return to text.
- Nowadays, we count someone on the eve of their 21st birthday as being 20. We do not know how the Biblical figures were reckoned, but if a person’s age was given by their nearest birthday, the imprecisions in a chronological calculation would tend to be self-cancelling during the course of a list of many names. If age was reckoned as we calculate, the effect would be cumulative, so that a simple addition of the ages in the genealogies would result in an underestimation of the age of the Earth by a few years. See: Lightfoot, J., 1654. Harmony of the New Testament, section 9. In: J. R. Pitman (ed.), 1822. Lightfoot’s Works, Vol. 3. Return to text.
- This is how it was understood by the Septuagint. Return to text.
- Norman, T. and Setterfield, B., 1987. Geological Time and Scriptural Chronology, Supplement to The Atomic Constants, Light and Time, Stanford Research Institute International. Return to text.
- Whitelaw, Ref. 10. As far as textual matters are concerned Whitelaw and Setterfield make too many inaccurate statements to be relied upon. Return to text.
- The Codex Vaticanus, which for most parts of the Old Testament is the oldest extant Septuagint manuscript, is not extant in Genesis till 46:28. It has been historically the most important Seputagint manuscript, and has been the basis of the text from which there has been the most comment on the Septuagint in the past 400 years. See: Greenslade, S.L. (ed), 1963. The Cambridge History of the Bible: The West from the Reformation to the Present Day, Cambridge University Press, Cambridge, p. 208. The value of the type of text Codex Vaticanus has for the New Testament is a well known and continuing controversy among Bible believers. The validity of the arguments here presented does not depend on any particular view of the New Testament text discussion. Return to text.
- Some authorities, including the corrector of Codex Alexandrinus, have 187 and 782 for Methuselah’s dates, just like the Masoretic Text. This is usually viewed as a secondary development within the Septuagint to avoid the problem of the standard Septuagint figures, where Methuselah seems to survive the Flood. Return to text.
- The figures for the remainder of Noah’s life in the three texts are inferred from the age given at Noah’s death in the three texts at Genesis 9:29. Return to text.
- The figures for the remainder of Terah’s life in the three texts are inferred from the ages given at Terah’s death in the three texts at Genesis 11:32. Return to text.
- Gooding, D.W., 1976. Relics of Ancient Exegesis, Cambridge University Press, Cambridge. Return to text.
- The use of the present tense ‘he reigns’ rather than the past tense ‘he reigned’ has been established as one of the criteria for identifying the earliest form of the Septuagint of I Kings (called the ‘Old Greek’) as opposed to a later revision (called by some the ‘kaige recension’) which most of the Septuagint of I Kings seems to have undergone. For this issue, see: Barthélemy, D., 1963. Les Devanciers D’Aquila, Vetus Testamentum Supplements, Brill, Leiden. Return to text.
- Shenkel, J. D., 1968. Chronology and Recensional Development in the Greek Text of Kings, Harvard Semitic Monographs no. 1, Harvard University Press, Cambridge, Massachusetts. Return to text.
- Gooding, D. W., 1970. Review of Shenkel, Ref. 25, in the Journal of Theological Studies, 21:118–131. Return to text.
- See also the criticism of Shenkel in: Thiele, E. R., 1974. Coregencies and overlapping reigns among the Hebrew Kings. Journal of Biblical Literature, 93:176–190. Return to text.
- Gooding, D. W., 1965. Pedantic timetabling in the 3rd Book of Reigns. Vetus Testamentum, 15:325–335. Return to text.
- The Old Testament Peshitta probably dates from the first two centuries AD, although little is certain about its dating. Return to text.
- Other examples of analogy include the following: In Genesis 11 it is said of each patriarch in the genealogy in the Masoretic Text, ‘And he begat sons and daughters’, whereafter the Septuagint on each occasion adds, ‘And he died’. The Septuagint has assimilated the formulae to those of Genesis 5. The Samaritan Pentateuch goes a stage further and has given the age of each individual at death as well (just as in Genesis 5). In Genesis 7:11 the Septuagint has substituted the ‘27th day’ of the month for the Masoretic Text’s ‘17th day’ of the month by assimilation to Genesis 8:14. The Septuagint thus makes the Flood to last an exact calendar year. For a fuller list of chronological alterations by the Septuagint, see: Larsson, G., 1983. The chronology of the Pentateuch: a comparison of the Masoretic Text and Septuagint. Journal of Biblical Literature, 102/3:401–409. Return to text.
- This term does not imply that the texts originally had anything to do with the Samaritans, merely that texts of their type were the basis of the revision which flourished among the Samaritans. Return to text.
- Purvis, J.D., 1968. The Samaritan Pentateuch and the Origin of the Samaritan Sect, Harvard Semitic Monographs no. 2, Harvard University Press, Cambridge, Massachusetts, pp. 69–86. Return to text.
- Waltke, B.K., 1970. The Samaritan Pentateuch and the text of the Old Testament. In: New Perspectives on the Old Testament, J. Barton Payne (ed.), Word Books, Waco, Texas, pp. 212–239. Return to text.
- Purvis, Ref. 32, pp. 71–72, 86–87. Return to text.
- They are not in B (Codex Vaticanus) though they are in A (Codex Alexandrinus). Codex A, for most books, is regarded as showing a later form of the text than Codex B. The further division of witnesses suggests unreliability in the Septuagint tradition. Return to text.
- Waltke, B.K., 1965. Prolegomenon to the Samaritan Pentateuch, Ph.D. Thesis, Harvard University. Return to text.
- Waltke, Ref. 33, especially p. 234. Return to text.
- Bruce, F.F., 1982. Commentary on Galatians, New International Greek Testament Commentary, Paternoster, Exeter, p. 173. Return to text.
- Bruce also holds that this short chronology is in accord with the Jewish source Seder ’Olam 3. But this source always adopts the lesser of two numbers if it deems chronology ambiguous. See the Jewish Encyclopaedia, 1905. Funk and Wagnalls Company, New York and London, under ‘Seder ’Olam Rabbah’. Return to text.
- KJV-1611 reads: ‘Now the sojourning of the children of Israel, who dwelt in Egypt, was foure hundred and thirtie yeares.’ This translation renders the Hebrew asher as denoting the children of Israel, not the length of their sojourn. This is possible grammatically, though it makes the relative clause to contain less new information than it does in my rendering. Return to text.
- Codex A of the Septuagint has here the addition ‘they and their fathers’. Codex B has ‘435 years’ in the writing of its original scribe, but this has later been deleted. Again the division of Greek witnesses suggests their unreliability. Return to text.
- I would like to thank Dr D.W. Gooding for a personal communication in which he pointed this line of argument out to me. Return to text.
- Keefe, E., 1980. The Duration of Israel’s Sojourn in Egypt, M.Th. Thesis, Grand Rapids Baptist Theological Seminary. I AM grateful to Dr D. A. Carson for notifying me of the existence of this thesis and for providing me with a copy. Return to text.
- Exodus 2:24, 6:8, 32:13, 33:1, Leviticus 26:42, Numbers 32:11, Deuteronomy 1:8, 6:10, 9:5, 30:20, 34:4 etc. Return to text.
- Or rather, he is following a Greek Pentateuch with a reading now only preserved by the Samaritan Pentateuch and Philo. See: Kahle, P. E., 1947. The Cairo Geniza, The Schweich Lectures of the British Academy, Oxford University Press, London, pp. 143–145. Kahle likewise thinks that the plural ‘fathers’ rather than singular ‘father’ in Acts 7:32 is a Samaritan type reading, loc. cit. Return to text.
- Genesis 12:4. Return to text.
- Genesis 9:24 says that Ham was the youngest of Noah’s sons. The two triads of names (Genesis 5:32, 11:27) are given in the order of decreasing importance of the role each individual has in the subsequent narrative. It is also implied that Shem was not the oldest of Noah’s sons by comparing Genesis 11:10 with Genesis 5:32 and Genesis 7:6. The Samaritan Pentateuch failed to recognise that Genesis 11:27 need not speak of triplets. This mistake, combined with its inference that Terah had died when Abram left Haran, created a contradiction which it resolved by altering the text. Genesis 10:21 gives similar information. The phraseology giving the age of Noah at the time of the birth of his sons is slightly different from the phraseology for the patriarchs who precede him in Genesis 5. Whereas for the others it says ‘and X lived Y years and begat Z’, for Noah in 5:32 we read ‘And Noah was a son of 500 years and Noah begat Shem, Ham and Japheth’. The phrase ‘son of Y years’ is a common Hebrew way of giving someone’s age, but may be used here to indicate a less definite connection between the actual age and the birth of Noah’s three sons than the formula ‘he lived 500 years and begat …’ might have done. In addition the formula of Genesis 5:32 is different because it repeats Noah’s name saying ‘Noah begat’ rather than simply ‘he begat’. Return to text.
- Sarah is also Terah’s daughter (or female descendant?) as appears from Genesis 20:12, though she had a different mother from Abraham. It is not certain whether Terah had two wives simultaneously or consecutively. Sarah is 10 years Abraham’s junior (Genesis 17:17), implying that Terah had the two wives within 10 years of each other. Return to text.
- This name is spelled differently in Hebrew from the name of the son of Terah. Return to text.
- Against the view that it can be a pluperfect, see: Jouon, P. 1947. Grammaire de l’Hebreu Biblique, 2nd edition, Institut Biblique Pontifical, Rome, p. 322, note 2. Return to text.
- It seems to me that there is good evidence for the view that the verbal form here can be a pluperfect. See: Waltke, B.K. And O’Connor, M., 1990. An Introduction to Biblical Hebrew Syntax, Eisenbrauns, Winona Lake, Indiana, pp. 552–553. Return to text.
- Barr, Ref. 11, pp. 585–586. Return to text.
- See, for example: Lightfoot, J., Jan 28, 1649–1650. Harmony of the Four Evangelists, Part 3, pp. 138–139. In: Pitman, J.R., Lightfoot’s Works, Vol. 5. Return to text.
- For a possible solution, see the Letter to the Editor, this issue, pp. 39–40—Editor. [This letter and others are summarized in Is Cainan a gap?]. Return to text.
- That is the Apocrypha (for example, 4th Ezra 10:45), ‘Deuterocanonical’ literature (for example, Jubilees, Testament of Levi), the Talmud, the church fathers (for example, Eusebius), and the Targums (for example, Pseudo-Jonathan on Exodus 12:40) etc. Return to text.