Home » Kiosk » Kiosk Article » The Missing Night of Matthew 12:40

The Missing Night of Matthew 12:40

Glenn Miller is a well-known Internet apologist who specializes in answering “tuff” questions from his readers about the Bible. Miller began his website, A Christian Thinktank, in 1994 and has since added a considerable amount of material to it by addressing almost every conceivable objection made regarding Bible discrepancies, problems, and “alleged” contradictions. He is extremely popular and as of March 2003 has a backlog of over 2,500 questions waiting in his mailbox to be answered.

Since Miller is often quoted by inerrantists in Usenet discussions, I decided to critique his answer to the problem of the “Missing Night,” i.e., “the three days and three nights” that Jesus declared he would be in “the heart of the earth” (Matthew 12:38-40).

Let’s begin by looking at the problem as posed by a reader of Miller’s Website.

Reader’s QUESTION: This comes from a list of reasons why humanists don’t believe the Bible. I picked up their document from America On-Line. It is also a question I’ve had, but have never had answered. Jesus said he was going to be in the earth (buried) for three days and three nights. If he died on Friday and rose on Sunday morning, how is this three days and three nights.

Glenn Miller’s ANSWER:

This is one of the easier ones…the Jews counted PART of a day or night as a WHOLE day or nite, so part of Friday, all of Sat, part of Sun would be ‘three days and three nights’–it was a Hebrew idiom of the day….

David Lee writes: At this point Miller make’s a grievous error. The problem with his answer is it simply isn’t true. There was no such Jewish custom of counting any part of a day as a full day AND a night in the Bible or the Talmud. Miller has been deceived by his sources. Miller cites the same references which Dr. Harold Hoehner, Josh McDowell, and others cite in their works.

Miller continues:

We do the same thing of course…if I say I worked at the office all day, ‘all day’ Normally doesn’t mean 24 hours…it means most of the daylight hours or whatever…

David Lee:This answer apparently did not satisfy one reader who then sent in a question of his own about November 2001. The reader pointed out–correctly–that in our modern day usage of the word “day” in reference to the workday we obviously don’t mean a literal 24-hour day but a day equaling the normal period of time allotted to work by tradition [approximately eight hours]. He adds that if a cricket match was rained out we would say, “A whole day’s play was lost due to rain”… even though we knew that only six hours were actually lost and not a literal 24-hours. He then points out that if he said, “I stayed in America for three days and three nights,” the word “day” then would be used in the absolute sense. He asks Miller, “Thus, one day and one night should be a full circle. Don’t you think so?”

The questioner then presses Miller for his sources for his claim that “The Jews counted PART of a day or night as a WHOLE day or nite.” He finishes by writing that if we say “22 nights and 23 days” the word day does not include nights and…

Reader: Thus I think linguistically, “3 days and 3 nights” must include at least “three” full or part days and “three” full or part nights, separately. Three part days and only “two” nights, it seems, cannot be termed as “three days and three nights.” Please clarify.

Miller responds:

It is important to recognize first off, that the issue of “Don’t you think so?” needs to be answered definitively ‘no’…Idiomatic expressions in other cultures don’t have to make ANY sense to us at all. Our job as readers of the literature from another culture is to try to understand THEIR idioms, rather than judge them.

David Lee: I would like to call attention to the sleight of hand being used here. Miller has been misled by his sources. The burden is on the apologist to prove that the phrase in Matthew 12:40 is an idiom of speech. Just claiming it is an idiom doesn’t make it one. Apologists often use the “it was an ancient Ancient Near East (ANE) custom” defense or claim it was a “Hebraic idiom” and they often do it without citing any contemporary sources or parallel grammatical constructions from the Bible that would establish the claim as factual. This technique usually works because the average layperson is intimidated by the scholarship of the asserter and doesn’t bother to press the issue further. However, as we shall see, Miller does attempt to supply some sources for his claim (the same used by Hoehner, McDowell, and others). This makes him a cut above the average Usenet apologist, but alas, it isn’t sufficient to carry his claim.

Miller proposes to quote from three sources: the Old Testament, the Rabbinix, and the New Testament. He begins with the Old Testament.

Miller writes:

1. The OT data (to show that ‘on the third day’ = ‘after three days’)

Gen 42.16: “And he put them all in custody for three days. 18 On the third day, Joseph said to them, “Do this and you will live, for I fear God” and they are released ON that day (from the context of verses 25-26). In this case the ‘for three days’ meant only ‘into the third day.’ 1 Kings 20.29: “For seven days they camped opposite each other, and on the seventh day the battle was joined. ” In this case we have ‘for seven days’ meant only ‘into the seventh day.’

2 Chr 10.5: “And he said to them, ‘Return to me again in three days'” (NAS) with verse 12: “So Jeroboam and all the people came to Rehoboam on the third day as the king had directed, saying, ‘Return to me on the third day.'” In this case ‘in three days’ is equivalent to ‘on the third day.’

Lee writes: Neither of these quoted passages claim that any part of a day can be reckoned as a day AND a night so it is surprising that Miller would even cite them. Few dispute the fact that the Jews sometimes reckoned part of a civil day as a whole day for counting purposes–but not just any part of a civil day for a day AND a night. For example, if Joseph put his brothers in custody on Monday morning and released them on Wednesday afternoon, Wednesday would still be the “third day” although a little more than 48 hours had passed. Likewise, if he tossed his brothers in jail Sunday night at the beginning of day one, Wednesday afternoon would not only be day three but it would follow the third night. The typical Jew would not bother to say, “He put them in custody part of Monday, all day Tuesday, and released them when Wednesday was almost over.” This manner of speaking (or writing) is a little cumbersome and it would be much simpler to write, “he put them in custody three days.”

The problem arises whenever nights are combined with day. Then it clearly refers to a period of nighttime and a period of daytime, and not the generic 24-hour civil day. There are very few who have a beef with this (although the Wednesday-crucifixion theorists would, however, because they argue for a literal 72-hour period between the time of Jesus’ burial and resurrection).

Miller then cites two passages, often seen on the Usenet, that appear at first glance to be a plausible defense for the claim that “any part of a day was reckoned as a day AND a night.” The passages are nothing more than smoke and mirrors, as we shall see.

Miller first quotes a passage found in Esther 4:16-5:1. In Esther 4:16, Esther makes an appeal for Mordecai to “Go, gather together all the Jews who are in Susa, and fast for me. Do not eat or drink three days, NIGHT [emphasis mine] or day. I and my maids will fast as you do. When this done, I will go to the king, even though it is against the law. And if I perish, I perish.”

The fourth chapter ends with Mordecai doing as Esther had requested. Then, in the fifth chapter, first verse, it says, “On the third day, Esther put on her royal robes and stood in the inner court of the palace [a capital offense without the king’s invitation], in front of the king’s hall.” The chapter goes on to record that the king was pleased by Esther’s appearance and allowed Esther to approach him. Esther requested the King and Hamaan (an opponent of the Jews) to appear at a banquet she had prepared for both of them. Oddly, in the Hebrew it isn’t even certain that Esther actually broke her fast during this time although she did prepare the meal and was there while the meal was being served.

This passage is adduced by Miller (and countless other apologists) as proof that the Jews reckoned any part of a day as a day AND a night. However, this passage proves nothing of the sort. The apologist must assume–without any evidence–that Esther began her fast sometime during the daytime portion of day one and not the period of night preceding it. Moreover, it must be remembered that Esther asked the Jews to fast “three days, NIGHT and day.” Here we see the word “day” used in both of its senses, the three civil days, followed by the usage of day with night. Notice also Esther says to fast “NIGHT and day.” In the Hebrew text, the word “night” is used before “day” in this phrase (Hendrickson’s Interlinear). If the Jews began fasting, say, on Friday at sunset, the first night of the fast would have preceded the first day of the fast (the Jews began and ended their days at sunset). This means if Esther approached the king on the third day, (say Monday afternoon), then the third night preceded it and she would not only be approaching the king on the “third day” but after “three days and three nights” as well since the third night had preceded it.

Miller next turns to I Samuel 30:11-13 for support, but this passage, like the one adduced before it, fails to support what the apologists so desperately want it to: that any part of a day could be counted as a day AND a night. Let’s read the passage in context, beginning with verse 10 (KJV).

I Sam 30:10 “But David pursued, he and four hundred men: for two hundred abode behind, which were so faint that they could not go over the brook Besor. 11 And they found an Egyptian in the field, and brought him to David, and gave him bread, and he did eat; and they made him drink water; 12 And they gave him a piece of cake of figs, and two clusters of raisins: and when he had eaten, his spirit came again to him: for he had eaten no bread, nor drunk any water, three days and three nights. 13 And David said unto him, To whom belongest thou? and whence art thou? And he said, I am a young man of Egypt, servant to an Amalekite; and my master left me, because three days agone I fell sick.”

Miller concludes from this passage and the Esther passage before it:

Thus, the Old Testament gives the picture that the expressions ‘three days,’ ‘the third day,’ and ‘three days and three nights’ are used to signify the same period of time. [NT:CALC:73]

David Lee: Yes, the expressions “the third day” and “three days and three nights” CAN mean the same period of time. That is because if an activity began at nighttime, then on the daylight portion of the third day you would already have the third night precede it. But notice what they don’t show: that “three days and three nights” can be a period of less than three days and three nights. That is the foundation on which the apologist rests his case.

If the unfortunate Egyptian was found late Sunday afternoon (names of days are used for convenience only) and had not eaten for three days and three nights then we can deduce that he had not eaten from Thursday night around sunset. There would be three nights and three days in the count yet it would still be, in this case, the “third day” since the daytime portion of a Jewish civil day followed the night. In the Hebrew text it was “three days ago” from the day the Egyptian was found that he fell sick, but in the Septuagint (LXX) the day of his rescue was the “third day” since he fell ill. The difference is slight. The Hebrew text implies he fell sick three days before his rescue (hence sometime Thursday daytime in this scenario) and in the LXX, he fell ill and was rescued the “third day,” or about 48 hours later since the first day of his illness would be reckoned as the “first day.” Either way, the apologist cannot make a case for an idiom. If he adheres to the accepted Masoretic text, then the Egyptian would have fallen ill sometime during the day Thursday, was then abandoned, and then began fasting around sunset Thursday night. If he uses the LXX, then the Egyptian last ate Thursday around sunset and was abandoned by his master sometime between Thursday sunset and Friday sunset. Neither choice helps the apologist prove that “three days and three nights” is a period less than “three days and three nights.”


2. The Rabbinical literature also manifests this idiomatic range: Rabbi Eleazar ben Azariah, tenth in the descent from Ezra was very specific: “A day and a night are an Onah [‘a portion of time’] and the portion of an Onah is as the whole of it” [J.Talmud, Shabbath 9.3 and b.Talmud, Pesahim 4a]

Miller goes on to write:

This understanding was used in the numerous correlations between Jonah 1.17 (‘in the belly of the fish for three days and three nights’) and the OT passages cited above [e.g., Mid.Rabbath on Genesis 56 (on 22.4); Genesis 91.7 (on 42.17-18)].

David Lee: However, the passage from the Talmud does not support Miller’s claim. The “Onah” in the Talmud was a method of reckoning time for a woman’s sexual impurities following menstruation, childbirth, etc., and the Talmud made it clear elsewhere that a day and a night each were an Onah, and not an Onah as a combined unit. How apologists like McDowell and Geisler can continue to peddle this argument baffles me. Surely they would have read the rebuttals by now and amended their argument. But no, they continue to purvey this argument, hoping, I suspect, that their readers will not bother to check up on them and discover their misuse of a rabbinical term used for sexual connotations. The word “Onah” appears in the Old Testament only once (Exodus 21:10) and in that context refers to a man’s sexual obligations to his wife. In any case, it never is used as a substitute for the typical use of “day” or “night” as used by the common people.

Furthermore, Jesus declared it was “as Jonah was in the belly of the whale” that he would be “three days and three nights in the heart of the earth.” Since the Talmud quote postdated Jonah by about nine centuries, we have absolutely no evidence that the term “Onah” was used in his day to apply to either a day or a night. So we cannot argue that the “three days and three nights” that Jonah was in the belly of the whale was anything less than the expression indicates. Even if “Onah” could be used in place of a “day and night” why didn’t Jesus say he would be in the heart of the earth three Onah’s? This would have removed at least one difficulty. The bottom line is Jesus did not use the rabbinical term “Onah” but used the expression used by the common people in Palestine. Since by Josh McDowell and Norman Geisler’s own apologetics “any part of a Onah could be reckoned as a day AND a night” then Jesus cleared matters for us by defining how long his “Onah” would be. His specific term defined the length of his Onah. (Of course I realize that “Onah” doesn’t even apply here, I am using it for sake of argument.)

Miller goes on to cite Matthew 27:63, 64 as an example of a New Testament “proof” that such an idiom existed in Jesus’ time. In verse 63 the Pharisees of Jerusalem tell Pilate, “Sir, we remember that while he was still alive that deceiver said, ‘After three days I will rise again.’ 64 So give the order for the tomb to be made secure until the third day.” Miller notes “after three days” was somehow equivalent to ‘until the third day’ (not ‘until the fourth day’). He then finished with the questioner as follows:

This data should demonstrate the rough equivalence of the NT phrases.

At this time Miller introduces yet another objector to his apologetics with the following:

I have been scanning your web site and find it very informative. Your treatment of others and their opinions is to be highly commended. I do find that your attention to detail is very valuable but at times the original question seems to get lost. For instance you say that the old issue regarding the passage in Matt. 12:40 (3 days and 3 nights), is an easy one. Your supporting documentation is to the point. But your references do NOT answer the original question. Jesus arose in the morning (Mark 16:9), if we use the logic presented by you and others, we still have a problem…3 days and 2 nights. Jesus ‘was risen early the first day of the week.’ I can understand a partial day and a partial night as counting for a day and night…but if Jesus arose from the dead early Sunday then we have a missing night.

Can you help me out with this?

Miller replies as follows:

Thanks for your kind words…and for pointing out the need to make a clearer conclusion to the article (hopefully i can do that soon)….

As for your question,

1. the day started at sundown (as the sabbath does today in Israel), and ANY part of the night/day cycle counted for the whole (as the article pointed out).

David Lee: Which we have seen is incorrect.


2.Jesus’ death on Friday afternoon would have been part of the Thursday nite/Friday daylight “day.”

3.thus, we have THREE ‘day/night’ days involved: “thur nite/friday
daylight,” “friday nite/sat daylight,” “sat night/sunday daylight”
(remembering again that a part of a period counted for the WHOLE)

Miller’s reader responded: Interesting logic but cannot agree with it. Its your use of Thursday night
that is troubling.

Miller fired back:

Its common usage even today…if you have a block about it, just think about how Israel does it TODAY…

The sabbath runs from Friday sundown to Saturday Sundown…any point in time between those two is considered “on the sabbath”…and the two “halves” are NOT considered separate at all (the night before the dawn is NOT considered any different than the bright noonday hour).

David Lee: Sure it is. Why would the phrase “X days and X nights” be found in the bible if there were no difference? Furthermore, while we all agree that the nighttime portion of the Sabbath is as equal in its prohibitions as that of the daytime Sabbath, how does that prove that day and night are not separable? Our Wednesday night is part of our Wednesday day and are inseparable as far as the day of the week is concerned, yet that doesn’t mean that there is no distinction between Wednesday daytime and Wednesday nighttime. There is–especially if we were reading an account and the writer wanted to clarify the part of a day an action took place.

Miller continues:

You just have to go with the normal levels of ordinary precision…for example, for someone to say Jesus was mistaken when he said ‘are there not 12 hours in the day?’ when there are NEVER exactly 12 hours in a day is applying a false standard to ordinary discourse…at mathematical usage levels, ‘3 days and 3 nights’ COULD NEVER EXACTLY EQUAL ‘ON the third day.’

David Lee: Sure it could. If the event that started the count occurred on the nighttime portion of day one, then on the daytime portion of day X we would have as many nights as days in the count.


But they used it that way in common discourse ANYWAY (and we do the same in other areas ourselves)…maybe you are applying an inappropriate precision grid onto ordinary language?–its a common problem for people of all persuasions and belief, and one i constantly have to be on guard against myself (as a westerner and science-type)….

David Lee: The old “We are Westerners” argument, a variation of “We shouldn’t question Jewish idioms” or “We shouldn’t question ANE customs.” Which sounds impressive, because it is true, but Miller fails to show it applies in this case. Instead, he assumes it and then looks for arguments to defend it. The fact is, Miller fails to demonstrate that the troublesome passage from Matthew 12:38-40 is, in fact, an idiom. The problem stands and he needs to try again.


I dont know if this helps any, or just adds to the confusion..but i thought i would try again quickly before getting back to work….

David Lee: When Jesus mentioned there are 12 hours a day, he was not referring to our modern concept of 60-minute hours, but 12 equal divisions of time that a day was divided into. Summer days were longer than winter days, but each day was still divided into 12 equal divisions of time. Still, I agree that a day did not have to be a full 12 hours to be reckoned as a day in Jewish reckoning. If a child was born three hours before sunset, that day was reckoned as a day for counting purposes, and the next day would be day two, even when it was first beginning. There is no dispute by most skeptics that a part of a day at times was reckoned by the Jews as a day for counting purposes, but it is disputed that a part of a daytime period or a part of a nighttime period, even if very small, could each be reckoned as a day AND a night.

Editor’s Notes:
1) Miller permits the use of any and all material from his web site, A Christian Thinktank, stating that it is “prayerware.”
2) Miller’s web site is the source of quotes attributed to him here.