Click here to view CMI's position on climate change.
This article is from
Journal of Creation 28(2):89–96, August 2014

Browse our latest digital issue Subscribe

The biblical minimum and maximum age of the earth


by and 

Careful consultation with the biblical record reveals a series of timespans linking creation to the Crucifixion. As a thought experiment, we combined these timespans to estimate the minimum and maximum allowable date of creation. Our goal was not to contradict the existing body of literature on the subject, but to put constraints on what is and is not biblically allowable. Implied precision and potential cultural differences (e.g. calendar systems, birthday conventions, and rounding conventions) mean we cannot pinpoint the age of the earth to a single year, yet the accumulated imprecision from those sources is limited to a maximum range of 308 years. Even including textual variants and debates over interpretation does not allow for dates approaching 10,000, let alone billions of years of Earth history. Accounting for all presently known relevant details and assuming the Babylonian Captivity began in 587 or 586 BC, we can say with confidence that the Bible places limits on the year of creation between 5665 and 3822 BC. The uncertainty within this range is mainly driven by textual considerations. The Masoretic/LXX debate creates a 1,326-year dichotomy, the Long vs Short Sojourn positions differ by 215 years, and various interpretations of the lists of the kings of Judah and Israel equates to around 54 years of additional uncertainty. Christians should avoid dogmatic claims of dating precision greater than intended by the Bible that could be refuted with new evidence, causing some to mistakenly believe the Bible itself has been refuted. Yet the combined tally of all the available data gives us fairly tight constraints on the age of the earth.

Using the Bible to estimate the date of creation has a long and rich history. The early Church Fathers put numbers on it, as did scientific greats like Sir Isaac Newton (about 4000 BC) and Johannes Kepler (3992 BC). The great academic and Archbishop James Ussher’s date of ‘Oct 23, 4004 BC’ is perhaps the most famous estimated date in history, although he has been much maligned by scoffers in recent years.

Most scholars veered away from biblical fidelity in the 19th and 20th centuries and very few seemed interested in mining the Bible for chronological details. In more recent years, however, the pages of this journal have been filled with many contributions on the subject.

Starting with the first issue in 1984, Osgood1 began publishing a series of papers that stretched out over the next several issues, eliciting responses and counter responses from various people. Over the years, more than a dozen different authors have published papers on chronology in this journal. They disagree on some of the details and there have been several sharp disputes, but two things unite them: a belief in the perspicuity of Scripture and a desire to systematically derive a consistent biblical dating scheme.

We set out not to put a specific date on creation, but to put limits on the range of acceptable dates. And, while we certainly have strong opinions on how to resolve several of the historic debates, we wanted to know the ‘worst case’ scenario rather than to assume those opinions are correct.

We acknowledge the great amount of work that has already been done and we are indebted to the prior body of publication. However, there are several factors that have not yet been systematically outlined and these have a small but important effect on all date calculations. This paper was foreshadowed by one of the earlier contributors, Pete Williams, whose paper “Some remarks preliminary to a biblical chronology” appeared in these pages in 1998.2

Numerical locks

There are some specific dates given in the Bible that are not up for debate. When a biblical author says a person was X years old when something happened, if we do not take that as a historical statement we quickly get to the point where words have no meaning. Many such numbers can be found throughout the Bible. For instance, we know that Caleb was 40 years old when he was sent with the other spies to Canaan (Josh. 14:7), and we know that he was 85 when he approached Joshua after the invasion of Canaan was completed to request Hebron for his inheritance (Josh. 14:10). We also know that the spying was done in the fall because it occurred during the grape (and pomegranate) harvest (Num. 13:20, 23). Statements like these are a very important source of data for biblical chronology.

There are other statements which give us a span of time between events. For example, in the time of the Judges, the Ammonites attempted to lay claim to the Reubenite territory just south of Ammon and east of the Dead Sea. Jephthah taunted the Ammonite king, saying, “While Israel lived in Heshbon and its villages, and in Aroer and its villages, and in all the cities that are on the banks of the Arnon, 300 years, why did you not deliver them within that time?” (Judg. 11:26, ESV). Thus, it is clear that the Israelites had occupied that area for 300 years. This probably does not mean ‘exactly’ 300, but it proscribes any attempt to reduce the period of the Judges to much smaller values.

There are other numbers in the Bible, however, that are more ambiguous, and when we string together multiple dates and date ranges, each with a certain degree of built-in ambiguity, we must acknowledge certain limits to precision.

Factors which limit dating precision

To generate a potential range of dates for creation, there are several sources of imprecision for which we must account. Some of these sources are inherent in the way humans report numbers. Others come from ambiguous statements in the biblical text (such as Terah’s age at Abram’s birth, see below). Still others come from the fact that we do not know which time-keeping conventions the ancients may have used.

Williams used the phrase ‘cumulative imprecision’ to describe the problem.2 We will copy his terminology, but by ‘imprecision’ we do not mean ‘error’ or that the biblical authors were sloppy with their reporting. On the other hand, we should not read biblical time statements as though the intent of the authors was to build a minute-by-minute timeline of Earth history. Most of the time statements are simple reports of major happenings, and they tied those to a general series of datable events (like a man’s age at the birth of a son). Sometimes, but not always, a series of dates can be bridged by a spanning statement that reduces the cumulative imprecision. And considering that most dates are given in ‘years’, we should not consider these to be an exact day count. This is what we mean by ‘imprecision’.

Accounting for each source of imprecision widens the potential range of dates for creation, and there are many factors to consider, yet each source of imprecision has a limited effect. Therefore, the extent of the accumulated imprecision is also limited. We will consider each source of imprecision in turn.

Implied precision

When humans report measurements, the context or style of the report often implies the precision of that measurement. If someone were to claim a structure was 100 m long, but it turned out to be 101 m long, it would be false to claim the person said it was exactly 100 m long. One cannot arbitrarily change significant figures when reporting numbers. Another source of ambiguity deals with rounding of numbers, and we should not assume ancient writers used modern rounding conventions (e.g. anything ≥ .5 gets rounded up to the next integer). For all we know, they may have always rounded down.

An example can be found in 1 Kings 7:23 concerning the Bronze Sea Solomon commissioned to be made for the Temple: “it was round, ten cubits from brim to brim, and five cubits high, and a line of thirty cubits measured its circumference” (ESV). Modern scoffers often claim the Bible wrongly teaches the value of π (the circumference of a circle divided by its diameter) to be ‘3.0’ rather than the correct ‘3.14 … ’. They are claiming a greater precision than was specified. Ignoring whether the Bronze Sea was a perfect circle, and whether the diameter measurement was for inside or outside, it could be anywhere from 9.5 to 9.7 cubits in diameter (e.g. ‘10’) to give it a circumference of between 29.8 and 30.5 cubits (e.g. ‘30’) using the correct value of π. When our interpretation includes a correct understanding of implied precision, we find that the value of π derived from operational science agrees with the record of 1 Kings 7:23.

Calendar systems

In addition to the uncertainties generated by implied precision, one must also consider the time-keeping convention used by the people reporting those dates. Many ancient societies used lunisolar calendar systems, where months are tied to the lunar cycle, but an occasional 13th intercalary month is added to keep months aligned with seasons, since 12 lunar months are 11 days short of a solar year. Some societies also standardized the process with the addition of 7 deliberately placed intercalary months within 19-year cycles. This was more predictable than the ‘as needed’ method but still required an additional intercalary month every 80 years to keep months aligned with the seasons. However, standardization would often take centuries and different localities have often used conflicting systems. While we do not know the exact antediluvian method used, we do know that the Jews have used a lunisolar calendar since the Exodus, when Moses was directed by God to institute a new system (Exo. 23:16, Lev. 23:39, etc.).

There are abundant examples of cultures changing time conventions. Before Islam, the Arabs used a lunisolar system, but Muhammad arbitrarily abolished the use of intercalary months in the Qur’an (9:36–37).

Modern Muslim countries such as Saudi Arabia use a 12-lunar-month Hijri calendar, where a month in summer one year will be in winter 17 years later. Their year numbering starts with Mohammed’s move from Mecca to Medina in 622 AD. On the New Year’s day 2014 AD (1,392 solar years later), the Hijri year was 1435.

There are many other examples of societies wrestling with time measurement. For instance, the Romans arbitrarily changed the date of the New Year to January 1 in the second century BC. The ‘years of confusion’ which followed were resolved by the Julian calendar, which re-aligned the months to the seasons by having one year with 445 days. Many ancient cultures began their year at the vernal equinox, while others began at the autumnal equinox. Various European localities up to the Middle Ages used a diversity of days to begin the year after the Council of Tours outlawed New Year’s celebrations in 567. Even the Gregorian calendar system, with January 1 as New Year’s Day, was not adopted uniformly across Europe, with the British Empire holding out until 1752, and some jurisdictions even longer than that.

Ancient peoples living in temperate latitudes presumably measured tropical years instead of sidereal years.3 However, ancient peoples living near the equator (or in places with no pronounced seasonal differences, e.g. the way many people imagine the antediluvian world) might be expected to default to a sidereal year, when the sun/stars/Earth return to the same alignment. Since a sidereal year is only about 20 minutes longer (1 sidereal year = 1.00003878 tropical years), this would have no significant impact on any age calculations, to the nearest year, adding at most one hour every three years, or just under 14 days in 1,000 years. However, this would have affected Ussher’s ‘Oct 23, 4004 BC’ date, so the reader is cautioned.

The Mayans used multiple simultaneous calendars, including a 260-day divine calendar (the most important), a long-count calendar similar to the Julian Calendar (with which they dated past and future events), a civil calendar similar to the Gregorian calendar, and a 584-day calendar based on the position of Venus (where five Venusian years are about eight solar years, or 99 lunar months).

The point of this brief survey is to illustrate the fact that we do not know which convention was used in the ancient past, and we do not know if all biblical data are reported with the same convention.

Figure 1. Histogram of last digits of Patriarch age data reported in Genesis. Not included are any ages back-calculated from the text (e.g. age of Noah, Terah, and Jacob when Shem, Abram, and Joseph were born). Note that the only ‘9’ comes from Methuselah’s age of death, which may have been back-calculated in the original, for the year of his death was quite obvious and significant. This is clearly not a random distribution, but the final digits appear more random as time progresses. After the Flood, most digits appear and the distribution appears more or less random, with the exception of more zeros than expected.

Years may have been reported in systems other than ones that align with solar years, and multiple possible shifts of six months or more may have occurred when societies switched or reformed their calendar systems.

Cultural differences in birthday conventions and counting age

So far we have considered imprecision in number reporting and a diversity of changing calendar systems, but we must also consider how ages are reported. In some East Asian cultures, newborns are traditionally said to be 1 year old (better translated ‘in his first year’) and ages are advanced at the lunar New Year, instead of on the birthday. It is possible that a child in these cultures could be ‘2 years old’, while native English speakers would say ‘1 month’. In addition, some cultures count age from conception rather than birth.

People sometimes keep track of multiple time conventions simultaneously and can flip from one to the other at will, meaning it is often difficult for an outsider to keep up, and context is of utmost importance. Therefore, we must allow for two fewer years than the reported biblical ‘ages’ in order to account for unknown birthday conventions.

Rounding imprecision accumulates

There are detailed genealogical lists in the Bible (e.g. Matt. 1, Luke 3). Some, however, come with specific dates and ages (e.g. Gen. 5, Gen. 11). The latter are more properly called ‘chronogenealogies’4 and they are of utmost importance, for they allow us to build a straightforward history of the time period they cover. Yet, there are certain facts about these numbers for which we must account. The chronogenealogies of Genesis are not based on a calendar system. The years are given as the age of the father, not the age of the earth (anno mundi or AM). If, as in modern English-speaking cultures, they used zero-based ages incrementing on birthdates, since a child can be born anywhere within that one-year span, each generation should add an average of six months to the calculations. It is unlikely that a series of children would all be born on each successive father’s birthday or on the day before those birthdays.2 But, accounting for the possibility of both extremes allows us to better estimate the range of dates for creation. Assuming random birthdates and that the ages were zero-based, 10 generations would carry about 5 extra years. But if ages were one-based (babies are in their ‘1st year’; Gen. 7:6 and 11 hint this was their convention), we should subtract about 5 years for every 10 generations instead. Many scholars of the past, including Ussher,5 have failed to recognize what we call ‘date slippage’.

Table 1. Minimum (Min), maximum (Max), and simple additive (Add) dates for Adam to Noah from Genesis 5, accounting for potential differences in birthday and rounding convention. Additive dates were generated by simply adding up the given numbers in the text. Minimum dates take into account potential rounding and the possibility of a 1-based birthday convention. Maximum dates take into account the possibility of a ratcheting scheme with a 0-based birthday convention. See text for an explanation of the adjustment values at each generation.

To test the effects of date slippage over the number of reported generations between Adam and Noah, we created a simple Excel spreadsheet and populated it with pseudo-random numbers representing the month of birth of consecutive children over 10 generations. After 1,000 trials, fully 92% of the replicates (nearly 2 standard deviations) had a total slippage of 4–6 years and only 1.5% had a slippage of as few as 2 or as many as 8 years. This works for both positive (0-based) and negative (1-based) date slippage. Clearly, this is a factor that needs to be taken into account when attempting to date creation, but it primarily applies to the pre-Exodus chronogenealogies.

Rounding of ages?

Consider the first five biblical Patriarchs listed in Genesis 5. Their ages at the birth of the next in line and at death are listed, but nine of the ten ages end in a 0 or a 5 (figure 1). This suggests the numbers may have been rounded to the nearest five. Or they may have used a 5-year ratcheting scale, where the age was only incremented every five years, meaning Adam could have been nearly 135 and still truthfully report his age as ‘130’. The lone ‘2’ is Seth’s age at death. From Jared to Shem, we see two additional digits, giving the appearance that they rounded down and to the nearest ‘2’. The lone ‘9’ is Methuselah’s age at death. Interestingly, in both cases the distribution of the reported numbers is evenly balanced (i.e. about the same number of zeros and fives from Adam to Mahalaleel and about the same number of zeros, twos, fives, and sevens from Jared to Shem).6

We are not trying to prove these dates are rounded or ratcheted, but since the numbers are so odd (i.e. not what one would expect from a random sampling, as even the post Flood patriarchs have three times more zeros than expected), we must allow for the possibility. In order to account for potential changes in rounding conventions, we will allow for a 5-year rounding convention from Adam to Mahalaleel, a 2–3-year rounding convention from Jared to Shem (table 1), and 1-year rounding (i.e. the modern convention) after that (table 2).

Why might the author of this section of Genesis have rounded these ages to the nearest five years? Possibly this was due to their great age, where a count precise to a single year might not be all that important to the individual when reporting his age, although ratcheting is more likely in this case. Searching for a mathematical reason for the apparent rounding leads us to consider the possibility that the first few generations measured ages in 60-month periods.

Initially, the lunar cycle would have been the most obvious way to track time, especially if Eden and/or its environs did not have significant seasonal variance.

They may have measured longer periods of time in groups of lunar months instead of years. If the first five patriarchs reported ages in 60-month blocks, the ages may have been converted later by multiplying by five, giving us the ages we have in the biblical record (with one exception) in 12-lunar-month years. There are many possible reasons for the appearance of these numbers, including random chance, but we are obliged to consider both rounding and ratcheting in our calculations because we cannot rule out these possibilities.

Table 2. Minimum (Min), maximum (Max), and simple additive (Add) dates for Arphaxad to Jacob from Genesis 11, 21, and 25, accounting for similar potential differences in birthday and rounding convention as in table 1.

Calculating the timespan and range

The above imprecision factors come in two categories: ‘per-link’ and ‘overall’. The following calculations will accumulate per-link factors (such as from birthday conventions & rounding), then apply the overall factors (such as calendar conventions) at the end.

Creation to Noah

Table 1 lists the minimum, maximum, and simple additive dates for Adam to Noah from Genesis 5, accounting for potential differences in birthday and rounding convention.

Noah to Arphaxad

Genesis 7:6 and 7:11 state the Flood started in Noah’s 600th year, and 8:13 states the Flood ended in his 601st year. This eliminates the possibility of ±5 rounding. Applying the limits of potential birthday conventions and offsets, we find the Flood started between 598 and 601 years after Noah’s birth. The simple additive date for the Flood is AM 1656, but it could have been anywhere from AM 1626 to AM 1693. Genesis 11:10 tells us Arphaxad was born two years after the Flood. This could mean ‘in the second year after the Flood started’ (just over one year after the Flood ended), ‘during the second summer/winter/fall/spring after the Flood ended’, or up to not quite 3 years after the Flood ended.2 Simply adding up the spans shows Arphaxad was born around AM 1659, with an outside range of 1628 to 1697. Note that we skipped Shem on purpose, because the best links are from Noah to the Flood to Arphaxad, making the ambiguity of Shem’s birth year irrelevant.

Arphaxad to Terah

Table 2 lists the minimum, maximum, and simple additive dates for Arphaxad to Jacob from Genesis 11, 21, and 25.

A 50-year ambiguity from Terah to Abram

The age of Terah when his son Abram was born is ambiguous because we only know Terah was 70 when Abram’s oldest brother was born. The narrative from Genesis 11:26–12:5 states that Terah, Abram and family moved from Ur to Haran, lived there a while, and Abram moved on from there to Canaan. That narrative implies (and Acts 7:4 confirms) Abram waited until his father died before leaving for Canaan, and states he was 75 when he left. If Abram left very soon after Terah died at 205, this would have made Terah 130 when Abram was born. But the text does not exclude the possibility that Abram waited.2 He may have lived in Haran for decades after his father Terah died before leaving for Canaan. All we know is he was old enough to be married to a wife 10 years younger (Gen. 17:17) before (Gen. 11:31) they moved to Haran. Terah therefore may have been as old as 180 when Abram was born, assuming Sarai was at least 15 when she married Abram. This is a break from the strict chronogenealogy and impacts the date of creation by up to 50 years.

Abraham to the Exodus

Genesis 21:5, 25:26 and 47:28 and Exodus 12:40–41 allow us to estimate the number of years from Abram’s birth to the Exodus. Assuming a plain reading of Exodus 12, this amounts to 720 years, 430 of which occur between Jacob’s move to Egypt and the Exodus. The 400 years of Gen. 15:14 would start in Exodus 1:8 when the Pharaoh who knew Joseph was replaced by one who enslaved the Israelites. Note that although Genesis 21:5 says Abraham was 100 when Isaac was born, this does not allow for ±5 rounding because in Genesis 17:1 we were told he was 99 the year before Isaac was born. Jacob and 11 of his sons moved to Egypt in AM 2299. Simply adding the spans puts the Exodus in AM 2729 with a range of 2676 to 2834.

However, Ussher5 and others7,8,9,10,11 have proposed that the 430 years Israel lived in Egypt started with God’s promise to Abraham in Genesis 12:1–3 instead of with Jacob’s arrival in Egypt. The 400 years of Genesis 15:14 would then start in Genesis 21:8–9, when Ishmael mocked Isaac at his weaning feast.

Rather than attempt to resolve this historic debate here, we acknowledge both positions have strengths and weaknesses, and include the range from both positions for the range of the Exodus: AM 2461 to AM 2834. From this point on, we will use the timespan for the Long-Sojourn view, acknowledging the Short-Sojourn additive, minimum and maximum dates will be 215 years less.

The Exodus through the Babylonian captivity

The books of Kings and Chronicles contain an unbroken chain of timespans from the Exodus to the Babylonian captivity. Simply adding up the years with the maximum length within the implied precision from each link yields a maximum biblically compatible timespan of 437 years from the beginning of Solomon’s reign to the Babylonian Captivity.

Thiele12 claimed regnal years were reported by two different systems: ‘accession year’ (1-based) and ‘nonaccession year’ (0-based) reckoning. He presented evidence of swaps between conventions in both Judah and Israel, in addition to the two kingdoms using differing conventions simultaneously, which limits the precision of dating simply based on cross-referencing regnal years.

Further complicating the matter, Judah appears to have advanced regnal years in the spring (Nisan), when their new year began, while Israel advanced theirs in the fall (Tishri), when their new year began. Thiele’s 383 years from the start of Solomon’s reign to the Babylonian Captivity is probably the shortest timespan proposed by a conservative scholar. Additional modifications and discussions of Thiele’s work can be found in Kaiser13 and Kitchen.14 Jones15 accounts for changing regnal year conventions and differing new year months using a more straightforward interpretation than Thiele to arrive at a longer timespan of 429 years. Pierce16 rejects Thiele completely. And Clarke rejects Austin’s, and Ashton and Down’s,17 attempts at linking biblical chronology to Egyptian chronology because they base their ideas on Velikovsky, whom he claims has been thoroughly discredited.18 All of these authors have a high view of Scripture. Clearly, biblical chronology is a difficult subject.

Table 3. Differences between the Masoretic (Mas.), LXX, and Samaritan Pentateuch relevant to the date of Creation.

* Some English versions mistakenly translate Nahor’s age at Terah’s birth as 179 years old, but the Greek manuscripts read 79.

The Babylonian captivity to Christ

II Kings 23–24 states that the Kingdom of Judah was carried into captivity in three waves, and the extra-biblical historical consensus is that these waves occurred in 597 BC, 587–586 BC, and 582 BC.

The only biblical timespan between then and the New Testament comes from Daniel 9:24–26. This prophecy places a minimum of 7 + 62 ‘sevens’, commonly assumed to mean 483 years from ‘the decree to rebuild Jerusalem’ until the Crucifixion of Jesus Christ. Yet, there are multiple such decrees, and we are not sure to which Daniel refers, although Austin argues strongly for one specifically, while at the same time removing a gap of 80–82 years, inserted by Ussher and others, by equating Darius to Artaxerxes.19

We must also rely on extra-biblical history to pinpoint the birth of Jesus Christ. This seems to be fairly well established at around 4 BC, although there are various biblically conservative counter arguments for a variety of dates in that range. The year of Christ’s death can be garnered from secular sources, and is attested by Daniel 9. Yet, we chose to peg our age estimate to the start of the Babylonian captivity because it allows for a slightly higher degree of certainty and because there is little dispute after that date.

Masoretic vs LXX vs Samaritan Pentateuch

A few hundred years before Christ, Alexandrian Jews produced a Greek translation of the Old Testament called the Septuagint (commonly abbreviated LXX). The authors of the New Testament frequently quoted directly from the LXX when referencing the Old Testament. The Masoretic text is the collection of Hebrew Scriptures collated around AD 700–1000 and is the basis of most modern Old Testament translations. We have many ancient fragments of Scripture in Hebrew (e.g. the Dead Sea Scrolls), which match the Masoretic very closely, showing the quality of work of the copiers in the intervening years, and supporting the authenticity of the Masoretic.

The LXX puts the earth significantly older than the Masoretic: including 586 additional years before the Flood and 780 additional years from the Flood to Abraham’s grandfather, Nahor (table 3). This is mostly due to the LXX including 100 more years in the ages of various Patriarchs at the birth of their son. The LXX also includes a Patriarch named Cainan between Arphaxad and Salah in Genesis 11:13.

This name does not appear at that point in the Masoretic or Samaritan Pentateuch. Most Greek texts of Luke 3:36 agree with the LXX on that point. From Terah forward, the primary date-relevant conflict is 1 Kings 6:1, in which the LXX dates the beginning of Solomon’s temple to 440 years after the Exodus vs 480 in the Masoretic. Even though we favour the Masoretic, we cannot know for certain, and therefore must acknowledge the possibility of the older dates from the LXX by adding 1,326 years to the maximum age allowed by the Masoretic.

There is another source of differing chronological data, the Samaritan Pentateuch. Written in Hebrew, but with a different etiology, it differs from the Masoretic in several thousand places, sometimes agreeing with the LXX and sometimes not.4 We do not put much stock in its authority, but see table 1 for details. It subtracts 349 years before the Flood and adds 650 years after it, for a net of 301 years more than the Masoretic.

Discussion and conclusions

Limited, gap-free imprecision

As detailed above, there are no chronological gaps from Genesis 1:1 to the Babylonian Exile. There is also no place where the text allows the insertion of an unlimited amount of time. In addition, this paper also takes the Genesis 1 narrative literally, leaving no room for a time gap there. Many have attempted to argue for gaps in the Genesis chronogenealogies, but, for example, even if Enoch were Jared’s great-grandson rather than his son, that would not change the timespan; Jared was still 162 when Enoch was born and this would not change the date of creation. Thus, there is no reason to argue for these gaps.

Table 4. Final Earth age range estimates (all dates BC).

* Minimum with 12-lunar-month years prior to the Exodus.

Ambiguities and imprecisions do not equate to falsehoods

The ambiguities detailed here do not mean the text is untruthful or erroneous. That a modern Western person would use a different number convention to describe age than someone of a different culture or time does not mean that either party is mistaken or lying. It merely means that a proper time convention translation is necessary. In the absence of complete information, the number should be understood to imply a range of possible ages. Our interpretation needs to allow for various possible implications of the original text, resulting in a range of possible ages. A range more narrow than intended by the Bible could conflict with valid outside evidence, and influence people to (incorrectly) disbelieve the Bible. But the Bible does make historical claims that can be used to estimate the age of the earth, so we should not pretend the earth could be any age. These claims can and should be used by Christians to evaluate the accuracy of extra-biblical historical claims.

Resulting date ranges

Table 5. Date estimates for the Flood (all dates BC).

* Minimum with 12-lunar-month years prior to the Exodus.

From creation to the Babylonian Captivity, we calculated a per-link imprecision of 219 years (including the 50-yr ambiguity concerning how long Abram remained in Haran), plus an overall systemic imprecision of 89 years. It is not possible to date creation with any more accuracy using just the genealogical data. We should allow for the possibility of ±10 years of imprecision from calendar system changes, and the possibility of up to 3% less solar years before the Exodus if the ancients used 12-lunar-month years or longer blocks of lunar months which would later be converted to 12-lunar-month years. We must also consider the possibility of 1326 additional years if the LXX chronogenealogies represent the original wording, 301 additional years if the Samaritan Pentateuch is correct, 215 less years for the ‘Short Sojourn’ view, and 46 fewer or 8 more years due to the ambiguities in the king lists of Judah and Israel. This yields an outside range of 3236 to 5078 years from Creation to the Babylonian Captivity. If the traditional historic date of 587 BC or 586 BC for the Captivity is correct, the earth cannot be more than 7,680 years old (table 4), having been created between 5665 BC and 3822 BC. The date of the Flood is more significant to the evaluation of extra-biblical history than is the date of creation. The Flood probably occurred between 2600 BC and 2300 BC, but certainly between 3386 BC and 2256 BC (table 5).

Note that the only way to get a ‘traditional’ date of creation of approximately 4000 BC is to use the Short Sojourn calculation and minimal to simple-additive adjustment parameters. This makes it likely that the earth is several hundred years older than most biblical creationists expect. However, we reject the idea that there are ‘missing generations’ that might increase the age to as much as 10,000 years, as Whitcomb and Morris did in their seminal and influential book The Genesis Flood.20


We thank L. Cosner and E. Killingsworth for many helpful suggestions after reviewing an earlier draft of this paper. We also thank the two anonymous reviewers for helping to hone our arguments while critically commenting on our work. Soli deo gloria.

LETTERS TO THE EDITOR Journal of Creation 28(3) 2014

The biblical minimum and maximum age of the earth

David Austin writes:

In the Journal of Creation article, ‘The biblical minimum and maximum age of the earth’, by Chris Hardy and Robert Carter, vol. 28(2):89–96, 2014, it states:

“And Clarke rejects Austin’s, and Ashton and Down’s, attempts at linking biblical chronology to Egyptian chronology because they base their ideas on Velikovsky, whom he claims has been thoroughly discredited.”

This statement concerning Clarke’s rejection of my biblical chronology because my ideas were supposedly based upon Velikovsky is decidedly wrong. I ask Hardy and Carter the question: “Where has Clarke claimed, in any one of his articles, that I have linked my biblical chronology to Egyptian chronology on the ideas of Velikovsky?” Surely the onus is on Hardy and Carter to reference even one article that makes this claim if their statement is to be believed. To my knowledge Clarke has not even seen, studied, or referred to my biblical chronology of the Old Testament (which is different, anyhow, from Velikovsky’s). Have Hardy and Carter studied Velikovsky’s chronology and compared it to mine?, In my first 3–4 articles that were published in Journal of Creation, there is not a shred of evidence that Velikovsky’s chronology ever even entered my mind, so how could I find a basis for ideas of an ‘unchanged’ chronology from an ‘unknown source’? Philip Mauro’s interpretation of the Scripture in The Wonders of Bible Chronology, also Martin Anstey’s The Romance of Bible Chronology, were the ‘basis of ideas’ for these articles, and many times I have replied to letters, articles, etc. from these sources.

In certain places in my articles I have agreed that Velikovsky’s dates of events have synchronized with my chronological dates of events, but these synchronizations were not the basis of my chronology.

I am asking, please, for Hardy and Carter, in reference to my biblical chronology, to completely retract their above statement if they cannot provide the requested evidence to support it.

David Austin
Tweed Head South,
New South Wales,

Robert Carter replies:

It was our intention to stay as far away from the middle of these disputes as possible while at the same time drawing a ring around the various parties. We did include a brief description of the state of the art and a summary of only a few of the various contentions concerning biblical chronology, however, and this is what Austin has focused on. While his letter has nothing to do with our methodology or conclusions, Austin does raise an objection that we need to answer.

In ‘The Queen of Sheba and the Ethiopian problem’,[1] Clarke discusses Austin’s views thus:

 “The central argument of David Austin’s 2012 Viewpoint in this Journal, ‘The Queen of the South’ is ‘the Queen of Egypt’[2] (hereafter QSQE)”.

Then, in the next paragraph:

“The core of the QSQE this: if Jesus’ ‘Queen of the South’ can be linked to Daniel’s ‘King of the North’ (Syria), and ‘King of the South’ (Egypt), the Queen of the South must be Egyptian also. It all sounds very straightforward, but as we shall see, linking this idea to the notion of this queen being monarch of Egypt and of a land or region called Æthiopia by Greek and Roman historians is not as simple as followers of Velikovsky’s revision would have us believe.”

And, in the third-to-last paragraph:

“If Velikovskians are as secure in their beliefs as they claim to be, they should easily be able to see off counter views such as mine without resorting to the sorts of fallacious and, at times, quite inappropriate tactics that, I have to say, have been rampant within the QSQE.”

These are direct references to Clarke thinking Austin has based his views (at least on the subject of the Queen of the South/Sheba/Egypt) on Velikovsky. But if on this one subject, and since Clarke directly accuses adherents to the QSQE of being ‘Velikovskian’, it is an easy extrapolation to link Austin’s entire biblical chronology to Velikovsky in the mind of Clarke. Thus, we have no need to retract anything and there is no reason to debate the matter here. This is an issue between Austin and Clarke and we would encourage Austin to take it up with him in these pages. We regret that we did not include this reference in our article as it would also have taken the onus off of us to prove why Clarke rejects Austin’s views on at least this matter.

Robert W. Carter


  1. Clarke, P., The Queen of Sheba and the Ethiopian problem, J. Creation 27(2):55–61, 2013; creation.com/images/pdfs/tj/j27_2/j27_2_55-61.pdf.
  2. Here he references Austin, D., ‘The Queen of the South’ is the ‘the Queen of Egypt’, J. Creation 26(3):79, 2012; creation.com/images/pdfs/tj/j26_3/j26_3_79-84.pdf.

References and notes

  1. Osgood, J., The times of the Judges, J. Creation (formerly CEN Technical J.) 1(1):141–158, 1984. Return to text.
  2. Williams, P., Some remarks preliminary to a biblical chronology, J. Creation (formerly CEN Tech J.) 2(1):98–106, 1998; creation.com/chronology. Return to text.
  3. Note that recent discussions on the length of the solar day do not come into play here. The length of the tropical and sidereal year are unaffected by the earth’s spin. See Carter, R.W., ‘Ancient’ coral growth layers, J. Creation 26(3):50–53, 2012; creation.com/ancient-coral; Jackson, D., Letter to the Editor; reply by Carter, R., J. Creation 28(2):64–65, 2014. Return to text.
  4. Sarfati, J., Biblical chronogenealogies, J. Creation (formerly TJ) 17(3):14–18, 2003; creation.com/biblical-chronogenealogies. Return to text.
  5. Ussher, J., Annales veteris testamenti, a prima mundi origine deducti, 1650, in which he calculated creation to have occurred on the “nightfall preceding 23 October 4004 BC”. A modern English translation, Annals of the World (Master Books, Green Forest, AR, 2006), is available from various sources, including creation.com. Return to text.
  6. One might be tempted to apply something like a simple Pearson Chi-Squared test to the data (Adam to Jacob). Doing so strongly rejects (p = 7 x 10–8, with 9 degrees of freedom) the null hypothesis that these numbers form a random distribution, but knowing that this is not a random distribution is not the same thing as understanding why the numbers are reported in this way. Return to text.
  7. Jones, F.N., The Chronology of the Old Testament, Master Books, Green Forest, AR, 1993. Return to text.
  8. Beechik, R., Sojourn of the Jews, J. Creation (formerly TJ) 15(1):60–61, 2001; creation.com/sojourn-of-the-jews. Return to text.
  9. Austin, D., Chronology of the 430 years of Exodus 12:40, J. Creation 21(1):67–68, 2007. Return to text.
  10. Austin, D., Reply to Brenton Minge, J. Creation 22(1):59–60, 2008. Return to text.
  11. Viccary, M., Reply to Brenton Minge, J. Creation 22(1):60, 2008. Return to text.
  12. Thiele, E.R., The Mysterious Numbers of the Hebrew Kings, Zondervan, Grand Rapids, MI, 1951.
    [note: An article that addresses independent evidence supporting Thiele’s dating method appeared online after the publication of the current article, see: Young, R.C., Evidence for inerrancy from a second unexpected source: the Jubilee and Sabbatical cycles, Bible and Spade 21(4), 2008;biblearchaeology.org/post/2015/07/25/Evidence-for-Inerrancy-from-a-Second-Unexpected-Source-The-Jubilee-and-Sabbatical-Cycles.aspx.] Return to text.
  13. Kaiser, W.C., A History of Israel from the Bronze Age through the Jewish Wars, Broadman and Holman, Nashville, TN, 1998. Return to text.
  14. Kitchen, K.A., On the Reliability of the Old Testament, Eerdmans, Grand Rapids/Cambridge, 2003. Return to text.
  15. Jones, F., Chronology of the Old Testament, King’s Word Press, The Woodlands, TX, 1999. Return to text.
  16. Pierce, L., Evidentialism—the Bible and Assyrian chronology, J. Creation (formerly TJ) 15(1):62–68, 2001. Return to text.
  17. Ashton, J. and Down, D., Unwrapping the Pharaohs, Master Books, Green Forest, AR, 2006. Return to text.
  18. Clarke, P., Why Pharaoh Hatshepsut is not to be equated to the Queen of Sheba, J. Creation 24(2):62–68, 2010; See also Habermehl’s letter to the editor and Clarke’s reply in J. Creation 25(1):44–46, 2011. Return to text.
  19. Austin, D., Is Darius, the king of Ezra 6:14–15, the same king as the Artaxerxes of Ezra 7:1?, J. Creation 22(2):46–52, 2008; creation.com/darius. Return to text.
  20. Whitcomb, J.C. and Morris, H.M., The Genesis Flood, Presbyterian and Reformed Publishing, Phillipsburg, NJ, 1961. Return to text.

Helpful Resources

Readers’ comments

Howard B.
Thank-you. Searching out what is true from and in God is always enjoyable!
Howard B.
Thank-you for this article. I think chronology is important, and very relevant to an understanding of history, which, of course, is governed by YHVH God.

Concerning the years of slavery in Egypt for Abraham's descendants, it would seem that Joseph being sold by his brothers as a slave and going to Egypt would be a reasonable starting point.
Robert Carter
One might want to start the Sojourn with Joseph's entry into Egypt, but one would still have to work out the relevant details, and this would not necessarily be easy. Chronology is important, but it is also difficult, which is why it is also enjoyable.
Leigh G.
Couple of points:

LXX can be immediately discounted as accurate in the chronological sense as when worked out, Methuselah would have had to have died 14 years AFTER the flood. I find this to be conclusive proof that the LXX is wrong when it comes to chronology.
Further corroboration comes from the fact that on the LXX chronology the division of the earth in Peleg's time is 536 years after the flood while in the Masoretic it is 101. If it was 536 then people indicated as being alive at the time would in fact be dead and therefore it simply cannot be correct.

Genesis 15 can be translated as "until" 400 years and also mentions the 4th generation. We also have a lot of information about the lives of the 3 patriarchs and their families right up to Moses from which we can work out quite detailed chronologies. I am convinced that the sojourn started when Abram was first called in Ur and left it, which would have been 20 years before he arrived in Canaan. A detailed examination (no room here) shows that Jacob entered Egypt in 1640BC approx, which is incidentally the decade of the eruption at Santorini thus explaining the worldwide famine.

The Lord himself was of course born in January 3BC, this date is concluded from comparing all available info especially Tiberius' reign, the dating of the temple to 46 years, Herod's death on 26th January 1BC, the Chinese supernova of April 4BC at his conception and the fact that Augustus instigated the first special census of the "entire world", which happened every 20 years or so for centuries, and that Quirinius held special office in the province of Syria for the specific purpose of carrying this assesment out in the region which included Palestine from January 3BC to December 2BC (Julian calendar).

I estimate AM as about 3938BC
Robert Carter

Concerning the LXX, please see the article mentioned in the prior comment (Textual traditions and biblical chronology). Concerning your other points, some are more or less valid but there is no way to interact at the level required in this comment section. If you would like to critique, add to, or point out an error in our paper, please send a letter to the editor of the Journal of Creation. One of the purposes of writing this paper was to spark creative conversation, so a response to our paper would be welcome. I note, however, and regardless of which historical pegs you chose to use to anchor your framework, that you seem to be using the simple additive method to calculate the date of creation. We showed that this is insufficient.
R. D.
One thing which persistently piques my curiosity is why, on the few occasions where this issue has been discussed, CMI authors never devote any time to the Samaritan Torah (SamTo). This piece mirrors the one other mention of it (in cited reference 2), in that it basically says "probably not much use" and leaves it at that.
From what I know of it, the SamTo is far more credible than the LXX and worth consideration to compliment and perhaps occasionally correct the Masoretic Torah, especially as it uses something closer to the original Kh'tabv Ibvriyt rather than the Masoretic's (later) Kh'tabv Ashuriyt - potentially significant given the ambiguities involved in ancient, pre-digit, systems to record numbers. The Genesis 5 genealogy in the LXX is self-contradictory, since it has Metushelah survive the flood even though he was not on the ark! The SamTo of Genesis 5 has Yared, Metushelah and Lamekh all die in the flood as unbelievers - possible but surely not very likely as Genesis 5:28-29 (which the SamTo includes) seems to regard Lamekh at least as a prophet. But the SamTo\LXX of Genesis 11 warrants serious consideration, surely? The figures corroborate each-other, and don't feature the Masoretic's massive para-flood discrepancy (where before sons don't appear until their fathers have lived several centuries, then afterwards are born at ages which are normal today; in the SamTo the postdiluvial ages come down but by far less).
This, as noted, leads to an extra 700 years between the flood and Abvraham, which is potentially very useful as it gives extra room into which to fit the start of civilisation. A pure Masoretic chronology requires civilisation to spring up in under two centuries (assuming the Peleg-Babel connection), which is an awfully short time.
Robert Carter
RD, you will be interested in a newer article by myself and Lita Cosner titled Textual traditions and biblical chronology. It appeared in the August 2015 edition of the Journal of Creation 29(2):99-105. In this article, we analyzed the Genesis 5 and 11 chronogenealogies of the LXX, the Samaritan Pentateuch, and the Masoretic. With one very minor exception, our conclusions strongly indicate that the Masoretic contains the original wording for these sections and that both the LXX and SP are deliberate recensions. It is not yet available online, so you can either wait for it, or subscribe to the Journal!
Stefan S.
I just would like to ask in which work Sir Isaac Newton addressed the question of the date of creation as 4000 BC. I could not find the source of such infromation even in Sarfati's 'The Great hoax on the Earth' - he mentions Newton in the sentence “Kepler calculated a Creation date of 3992 BC, and Isaac Newton (1643-1727), probably the greatest scientist of all time, also strongly defended biblical chronology” on the page 311. I am aware of Newton's statements like, for example “The Pentateuch is composed of the Law and the history of God’s people together; and the history hath been collected from several books, such as were the history of the Creation composed by Moses, Genesis 2:4" (Observations upon the Prophecies of Daniel, and the Apocalypse of St. John) and also Rhoda Rappaport mentions that Newton "vigorously opposed to a naturalistic cosmic evolution and suggested chemical processes that might aid in explaining the creation of the earth's crust during the six days" in 'When Geologists were historians' (1997, p.138), but I haven't found any paper or book by Newton yet that would specifically address the calculation of the time since creation.
Robert Carter

One source is Experimenting on a Small Planet: A Scholarly Entertainment by William W. Hay, Springer Science & Business Media, 2012, page 63 (available on Google Books). Perhaps this is not the most scholarly citation, and he does not provide a reference to the factoid himself. Since we cannot find it anywhere else, upon introspection we would like you to take Newton's 'date' with a grain of salt.
greg S.
it is interesting. i did the same thing but counted forward until i reached a known date for the destruction of jerusalem by king Nebuchadnezzar and the year Nebuchadnezzar started his reign. i call it after adam was created or AAWC. i used the years outlined in the bible and use scriptures. [link deleted per feedback rules] creation occurred in 3986 BC. year 1BC on my calendar counting from adam forward is 3986
Robert Carter
Interesting, but you used the 'simple additive' method. Also, 'AAWC' has traditionally been refered to as "AM" or anno mundi, the year of the world. Sorry that we could not include your link above.
Steve S.
Not that it affects any of the dates, but another way of looking at Daniel 9 is that the first seven weeks referred to the decree of Cyrus until Nehemiah, who got Jerusalem started again. Then after 62 weeks the anointing of the temple is cut off, and there shall be nothing left of it, for the people of the prince shall come and destroy it.

So the Jewish year of 5776, is it still okay to put the blame on bar Kokhba for the discrepancy between 6,000 years?

I still believe things are lining up to the creation week pattern so that the millennium is the sabbatical rest and sometime after that is the new creation. This is perhaps why we can't know the certain age.
Robert Carter
As in the first comment above, this is getting into eschatology, which we generally avoid. This is not because all of us do not have our personal opinions, but because most aspect go beyond our mandate. Please see End-times and Early-times and Why doesn’t CMI take a position on … ?
Deborah H.
I completely appreciate all of the intensely thinking minds and incalculable hours of research that have gone into excavating information within the Bible to get as close as possible to the age of the Earth. I am just a thinking Christian with no particular specialized academics -- just thinking. One thing that troubles me about all the calculating is that, to my understanding, we can have absolutely NO idea, even by reading the Genesis Creation story, how long Adam and Eve lived in the Garden of Eden BEFORE Lucifer deceived them and they were exiled from the Garden. I believe Adam and Eve never aged while in the Garden and were in the physical condition God originally made them in the entire time they lived there, since there was no death at that time. I don't believe God created them and within just a short amount of time they "fell." As far as we can possibly know, they could have lived there for thousands of years or for 10 years before the "fall." And, as always, we have the unknown time-frame between Genesis 1 verses 1 and 2 and 3. So, I don't know that even with all of the other obvious date calculations we can extract from the Bible that we can really determine with even close accuracy the age of the Earth, since those two very important factors will always remain unknown. I have personally never reach a conclusion on this issue, but I have an open mind. I do enjoy the interesting subjects examined on this website.
Robert Carter
Fascinating comment, but it calls for a response.

1) God commanded Adam and Eve to procreate. While one might suppose Eve may not have ovulated for a while, stretching that out to thousands of years begs the entire question.

2) If Adam and Eve waited a long time to commence the 'act', they would have been disobeying God. Thus, the Fall would not have happened a long time after Creation.

3) Adam was 130 years old when Seth was born. The birth of the son is pegged to the age of the father, not the time since the Fall. Also, the age of Adam is not in reference to the Fall. This constrains the time Adam and Eve could have remained in the Garden.

4) The time of the Fall (I think it was probably a few days or weeks at most, and you can find additional similar comments in various places on creation.com) is not involved in calculating the age of the earth, so the question, while interesting, is irrelevant.
Robert S.
This entire business is nonsense. It has been known for decades that the earth is approximately 4.5 billion years old. The age is now known to a precision of better than 0.01%: the actual age is 4,568,200,000 years [1].

Anyone who supposes that the bible is a historically accurate treatise is simply not paying attention to what science shows -- which is, of course, based on observations that anyone can make who cares to take the trouble to do so.

1. Bouvier, et al, The age of the Solar System redefined by the oldest Pb-Pb age of a meteoric inclusion. Nature Geoscience, vol. 3, Sept. 2010, p. 637.
Robert Carter
You present a figure with FIVE significant figures. That is amazing, because that is not the number I was presented with when I was in school, and I am certain it will not be the number a few years hence. If the number changes by greater than +/- 50,000 years, your claims of precision will be destroyed. Because you walked right into this newbie trap, I suspect you are not well trained in the sciences.

I note that you hurl some large stones, e.g. "anyone" "science" "observations" "take the trouble". But by doing so you show that you are 100% unaware of the information found on this website. You also are unaware of the backgrounds of the multiple PhD scientists on our team, nearly all of whom were at one time evolutionists, and many of whom were outright atheists.

You might also want to consult our radiometric dating Q&A.
Richard K.
I have tallied the number of years that Josephus gave in his history of the Jews and added it to the years since he wrote the book and it comes out to be over 7,000 years. Right around your high mark. I wish I had the number with me but I wonder why nobody has ever referenced his writings when talking about the age of the earth.
Robert Carter
One cannot say "nobody" has considered Josephus, but this article was specifically about the *biblical* age of the earth. Thus, we did not consult outside sources. Also, Josephus was strongly influenced by the LXX, hence his dates are in the high range.
Robert R.
Very interesting read. Just a simple theory to ponder. With God being a God of order, it is possible that the rapture would take place 5,593 years after creation. This would allow seven years for tribulation and 1000 year millennial reign to bring the total to 7,000 years. No scripture to support, just a thought.
Robert Carter
An interesting question, but CMI does not get into eschatology, specifically various pre-, post-, and a- viewpoints, so I will demure. Theologians have been discussing such things for many years. I believe it led to great millennial expectations 700 years after Christ's death, then again about the year 1000. Etc.

Comments are automatically closed 14 days after publication.