Skip to main content

Daf 3b - "In the middle of the night, in the face of death"

Dr. Moshe Simon-Shoshan

Ein Yaakov - The World of Talmudic Aggada

By Dr. Moshe Simon-Shoshan


Lecture 8:  Daf 3b

“In the middle of the night, in the face of death”



How many watches during the night?


The Gemara now returns to the question of how to break down the night.  As we have already seen, there were two options for breaking down the night in the ancient world:  three watches, which was the biblical (and Babylonian) tradition, or four watches, which was the Roman practice.  Previously, the Gemara sided with the threefold division.  Now, however, the Gemara presents these two options as a dispute between two Tannaim:


Our Rabbis taught:

The night has four watches.  These are the words of Rabbi. 

R. Natan says: Three. 

What is the reason of R. Natan?

It is written:

“So Gidon, and the hundred men that were with him,

came into the outermost part of the camp

in the beginning of the middle watch” (Shoftim 7:19).

And one taught:

Under 'middle' is to be understood

only something which is preceded by one and followed by one. 

And Rabbi? 

'The middle' means: one of the middle ones. 

And R. Natan?

Not 'one of the middle ones' is written,

but 'the middle' is written.


First the Gemara presents the dispute.  R. Natan holds that there are three watches, while Rabbi holds that there are four.  Interestingly, R. Natan, who is from Babylonia, holds that there are three parts to the night, as did the ancient Babylonians.  In contrast, Rabbi, AKA Rabbi Yehuda Ha-nasi, was one of the Patriarchs, who were known to be closer to Greco-Roman culture than their colleagues.  This may partially explain why Rabbi favors the Roman four-part division of the night.


The Gemara then presents R. Natan’s proof for his position.  As we discussed earlier, the verse in Shoftim refers to a “middle watch,” suggesting that there are an odd number of watches, minimally three.  This verse is quite a good proof that there are three, not four, watches.  What then does Rabbi do with this verse? He has little choice but to suggest an alternative reading of the verse.  The verse could mean only that the watch was one of the middle watches, leaving room for four watches.  R Natan responds that if this were the case, the verse should have explicitly stated that Gidon struck during “one of the middle” watches. 


The Gemara now turns to Rabbi’s proof for his position, which is not as straightforward as R. Natan’s:


R. Zerika, in the name of R. Ami, in the name of R. Yehoshua b. Levi, says:

One verse reads,

“At do I rise to give thanks unto Thee

because of Thy righteous ordinances” (Tehillim 119:62)

And another verse reads:

“Mine eyes forestall the watches” (Tehillim 119:148)

How is this?

[This is possible only if] the night has four watches.


Rabbi’s proof comes, not from a straightforward reading of a single verse, but from a deduction based on the juxtaposition of two verses from Tehillim 119.  Psalm 119, the longest chapter in Sefer Tehillim, is an eight-ply acrostic, consisting of eight verse units in which each verse starts with the same successive letter of the alphabet.  The entire chapter is devoted to praise of God and His word.  In the section of verses beginning with the letter chet, we read, “At midnight (chatzot) do I rise to give thanks unto Thee.” This verse seems to firmly establish the fact that David arose at midnight.  Yet later on in the psalm, in the tzadi section, the psalmist declares, “Mine eyes forestall the watches.” The exact meaning of this verse is clear.  Literally, the verse reads “my eyes precede (kidmu) the watches."  JPS translates the verse as, “My eyes greet each watch of the night.” Robert Alter, in his translation of Tehillim, similarly understands this verse as meaning that David stayed up all night studying Torah.  Amos Chakham z”l in Da’at Mikra, in contrast, understands the word ashmorot as singular, referring to the last, morning watch.  According to him, David arose before dawn, during the final watch.


The Sages, however, in keeping with midrashic method, present a hyper-literal interpretation of the verse.  They understand the words “my eyes precede” as meaning “I awake before.” They understand the term “watches” as referring to exactly two watches.  In other words, David arose two watches before the dawn.  Now, Rabbi argues, since we know from the first verse that David awoke at midnight, there must be two watches from midnight to dawn.  This means that there are a total of four watches. 


Just as the Gemara tried to understand how Rabbi would have responded to R. Natan’s proof that there are only three watches, now the Gemara tries to understand how R. Natan would have responded to Rabbi’s proof:


And R. Natan?

He is of the opinion of R. Yehoshua, as we have learnt:

R. Yehoshua says:

until the third hour, for such is the custom of kings, to rise in the third hour.

Six hours of the night and two hours of the day amount to two watches.

R. Ashi says: One watch and a half are also spoken of as 'watches.’


R. Natan has a different understanding of the second verse from Tehillim 119.  He reads the verse as saying “my eyes have preceded [other kings’] by two watches.” In other words, David arises at the time equivalent to two watches prior to the time when other kings awake.  If there are three watches, then each watch is four hours and David arises eight hours before the other kings.  If David awakes at midnight, as stated in the first verse from Tehillim, then this would mean that the other kings arise eight hours later, which takes us to the beginning of the third daylight hour.  R.  Yehoshua’s statement, that kings arise in the third hour, corroborates this finding.  Thus these two verses can be reconciled, even according to R. Natan’s position. 


R. Ashi has a more straightforward understanding.  According to his understanding, like Rabbi, R. Natan agrees that the verse refers to David awaking “watches” before the dawn.  However, R. Natan argues that the term “watches” does not necessarily refer to two or more watches; rather, it can refer to any amount of time more than one watch.  In this case, there are one and a half watches between midnight and dawn, according to R. Natan.  So R. Natan may read the verse as saying “I arise one and a half watches before dawn,” just as Rabbi reads it as saying, “I arise two watches before dawn.”


By citing these two verses from Tehillim in the discussion of the evening Shema, this section raises two larger thematic issues.  The first issue is the significance of midnight.  Previously, we saw that the Mishna concludes that midnight has no halakhic significance on the biblical level.  Biblically, the only significant points in the night are the end of the first watch and dawn.  When the Mishna states that a certain action must be done before midnight, it refers only to an artificial precaution instituted by the rabbis.  In fact, in all such cases, one can really do the action until the end of the night. 


Now midnight appears once again as central to Rabbi’s argument for a fourfold division of the night.  The fourfold scheme, by definition, divides the night in half and thereby marks midnight as a significant time.  The threefold model does not mark midnight at all.  Though the Bible mentions midnight, it gains real significance only in the Roman world.  The Romans, whose legacy lives on till this day, mark the beginning of the new day at midnight.  Not coincidentally, they are also the source of the fourfold division of the night.  The Gemara will continue its discussion of right after the next section.


These verses in Tehillim also raise the idea that the middle of the night is actually an optimal time for prayer.  Previously, we noted that there is no place for prayer between sunset and dawn in a model of prayer based on the Temple sacrifices, because the Temple did not operate during that time.  The idea of praying at night can be understood in two ways.  In an experiential model of prayer, all times are good for prayer.  Alternatively, prayer is particularly appropriate at night, when one may feel vulnerable and endangered. 


Silence of the Dead


The next small section is a digression from the Gemara’s focus on nighttime.  Since the Gemara just quoted a tradition transmitted by R. Zerika, in the name of R. Ami, in the name of R. Yehoshua b. Levi, the Gemara now presents another tradition that was handed down by the same line of scholars:


R. Zerika further said, in the name of R. Ami, in the name of R. Yehoshua b. Levi: One may discuss in the presence of a dead body only things relating to the dead. 

R. Aba b. Kahana says:

This refers only to matters of Torah, 

but as for worldly matters there is no harm. 

Another version is:

R. Aba b. Kahana says:

This refers even to matters of Torah. 

How much more so to worldly matters!


This discussion about appropriate speech in the presence of a dead body would appear to be linked to the well known ruling found in the Mishna at the beginning of the third chapter of Berakhot:







In both cases, we have rules that exempt or prohibit a person from doing normal activities in the presence of a dead body.  Why should these rules apply? If we focus on R. Zerika’s original statement, he appears to prohibit all speech not directly connected to the needs of the dead person.  In this case, we could suggest that the presence of the dead body obligates those around him to tend to his burial.  All other activities not connected to the burial, including speaking, are prohibited.  Another possibility involves the concept of lo’eg la-rash, “taunting the poor man.” Since the dead person has lost his ability to engage in human activities, such as speaking, doing so in his presence is disrespectful or even cruel.


R. Aba b. Kahana, in the first version of his statement, suggests that the prohibition against speech in the presence of the dead may be limited only to words of Torah.  All other speech would be permitted.  This prohibition does not seem to be motivated by an obligation to focus on burying the dead.  The second possibility we raised above, lo’eg la-rash, seems much more likely here.  In this view, the greatest loss to a dead person is the ability to do mitzvot.  Therefore, only words of Torah may not be spoken before the dead.


As we noted earlier, this passage appears here primarily because it is transmitted by the same chain of rabbis as the previous passage.  Otherwise, it appears to be a digression or non-sequitur.  However, I wonder if there is a thematic connection between this discussion of speaking before the dead and the surrounding passages.  In many forms of literature, death is linked with night.  Perhaps most famously, we have the images of the “shadow of death” in Tehillim 23.  If so, a discussion of prayer at night may be connected to a discussion of speech in the presence of death, especially words of Torah.


More Royal Insomnia


The Gemara now returns to its discussion of Tehillim 105 and its implications for King David’s nightly schedule:


But did David rise at ?

[Surely] he rose with the evening dusk?

For it is written:

“I rose with the neshef and cried,” (Tehillim 119:147)

And how do you know that this word neshef means the evening?

It is written:

“In the neshef, in the evening of the day,

in the blackness of night and the darkness!” (Mishlei 7:9)  

R. Oshaya, in the name of R. Acha, replies:

David said: never passed me by in my sleep. 

R. Zeira says:

Till he used to doze like a horse, 

from thence on he rose with the energy of a lion. 

R. Ashi says:

Till he studied the Torah,

from thence on he recited songs and praises.

But does neshef mean the evening?

Surely neshef means the morning?

For it is written:

And David slew them from the 'neshef' to the evening erev of the next day,

and does not this mean, from the 'morning dawn' to the evening?

No.  [It means:] from the [one] eventide to the [next] eventide. 

If so, let him write: From neshef to neshef, or from erev to erev?

Rather, said Raba:

There are two kinds of neshef:

[the morning neshef], when the evening disappears and the morning arrives,

[and the evening neshef], when the day disappears and the evening arrives.


Previously, the Gemara sought to reconcile two verses from Tehillim 119 concerning David’s nighttime activities: “At midnight do I rise to give thanks unto Thee because of Thy righteous ordinances” (62) and “Mine eyes forestall the watches” (148).  Now the Gemara seeks to reconcile this with yet another verse from the psalm, “I rose with the neshef and cried” (149).  The critical question remains, when is neshef? In biblical Hebrew, the word can refer to both dusk and dawn, much like the rabbinic term, bein ha-shemashot.  To which does it refer in this verse? Generally speaking, one speaks of rising before dawn, not before dusk.  Furthermore, as we have already seen, the very next verse says, “Mine eyes forestall the watches."  If, as Amos Chakham says, this refers to arising in the final watch of the night, then there is perfect parallelism between the two verses.  According to those who argue that this verse refers to staying up all night (see Radak), the interpretation of morning is still preferable.  We have then a classic example of intensification in biblical poetry.  The psalm first cites a more moderate image, “I arise very early!” and then a more extreme one, “I never really go to sleep!”


Interestingly, in attempting to clarify the meaning of neshef in this verse, the Gemara does not seek to read the verse in the context of the previous one, even though it has just discussed that same verse.  This illustrates how context is often irrelevant in midrashic interpretation.  In order to prove its point, the Gemara leaves Tehillim altogether to find a verse in Mishlei in which neshef refers to the evening.  Interestingly, in Mishlei, nighttime takes on a very different coloring than it did just previously in the verses from Tehillim.  Whereas in Tehillim, the night is a time for study and prayer, in Mishlei, the cover of night is used for an illicit sexual encounter.


The Gemara now cites three possible resolutions for the contradiction between the two verses – one says that David wakes up at dusk (i.e., he never went to sleep), and the other says that he wakes up at midnight.  R. Oshaya favors the verse which says that David stayed awake all night.  Therefore, he understands the verse about awakening at midnight in a slightly non-literal sense.  The Psalmist does not say “I arise” at midnight, but rather, “I am awake” at midnight, never having gone to sleep. 


R. Zeira takes a more diplomatic approach.  The two verses give conflicting accounts about whether David was awake or asleep from dusk until midnight.  He therefore suggests that David was somewhere between sleep and wakefulness at this time.  He “dozed” – he drifted on the periphery of sleep.  Hence, the descriptions of both verses are accurate.


Finally, R. Ashi suggests yet another solution.  Like R. Oshaya, he agrees that David never sleeps at all.  However, he envisions a greater significance to David being awake at midnight.  The various verses cited from Tehillim 119 do not clearly specify what David does when he is up in the middle of the night.  Does he pray, or does he study Torah? The relationship between prayer and Torah study appears throughout the first chapter of Berakhot.  Here, R. Ashi divides the night between the two.  David devotes the first half of the night to Torah, as we see from verse 147.  We understand this verse as referring to David being up all night, and its last words are, “I hope for Your word,” which is interpreted as Torah study.  Here, the context of the verse seems to be relevant to its interpretation. The previous verse, “Mine eyes forestall the watches…” concludes, “to dwell on Your utterances," which unambiguously refers to Torah study.


An image of King David emerges from this discussion who does not, or barely, sleeps, but, rather, devotes his nighttime to study and prayer.  David appears as a paradigm of commitment to spiritual life.


This website is constantly being improved. We would appreciate hearing from you. Questions and comments on the classes are welcome, as is help in tagging, categorizing, and creating brief summaries of the classes. Thank you for being part of the Torat Har Etzion community!