Dating the Death of Jesus: Memory and the Religious Imagination

Abstract

After discussing the scholarly preference for dating Jesus' crucifixion to 7 April 30 CE, this article argues that the precise date can no longer be recovered. All we can claim with any degree of historical certainty is that Jesus died some time around Passover (perhaps a week or so before the feast) between 29 and 34 CE. The emergence of the Johannine tradition (in which Jesus died on the day of Preparation) and the Markan tradition (in which Jesus died on the Passover itself) are explored through the lens of social/collective memory.

There is a relatively widespread consensus that Jesus was crucified on April 7, 30.¹

Jesus was dead by the evening of Friday, April 7, 30.²

As regards Jesus, it is certain that he was crucified in Jerusalem on 7 April 30 CE.³

the fourteenth of Nisan (7 April of the year AD 30) is, apparently in the opinion of the majority of contemporary scholars as well, far and away the most likely date of the crucifixion of Jesus.⁴

Thus, I conclude that Jesus died on Nisan 14 (April 7) in A.D. 30.⁵

As these quotations illustrate, 7 April 30 CE is widely regarded by many modern Jesus scholars as the day on which Jesus died.⁶ Of course, not everybody agrees with this view. Some favour the year 30 without specifying the precise day or month,⁷ others propose more idiosyncratic dates,⁸ and a sizeable minority of largely conservative scholars favour 33.⁹ Given the complexities of our sources, however, the level of agreement surrounding the date of Jesus' crucifixion is truly astounding. What I want to do in this article is (1) to ask how scholars got to 7 April 30 CE in the first place, (2) to look at the implications of this date, and (3) to suggest that scholarly confidence in it is severely misplaced.

1. Why 7 April 30 CE?

All attempts to date Jesus' death begin with the gospel passion narratives. As is well known, the Synoptic and Johannine accounts have both similarities and contradictions. Both agree on the general sequence of events: that Jesus ate a meal with his disciples on a Thursday evening, that he was crucified the following day and quickly buried before the approaching Sabbath, and that the tomb was found to be empty on the Sunday morning. The major chronological discrepancy between the two accounts lies in the precise way in which events map onto the Passover. In Mark, both Jesus' last meal and his death take place on the feast day itself, while in John both events take place on the day of Preparation (John 19.14; 18.28; 19.31).¹⁰ Thus both traditions agree that Jesus was crucified on the Friday and buried that afternoon; what is disputed is whether that Friday was the Passover itself, Nisan 15 (so Mark), or the day of Preparation, Nisan 14 (so John).

One approach to this problem is to harmonise the two traditions, to suggest that John and the Synoptics simply used different calendars. Perhaps the most famous exponent of this method was Annie Jaubert who, in a series of studies from the 1950s, famously suggested that Jesus and his disciples followed not the by now dominant Babylonian lunar calendar, but an old solar calendar evident in the books of Enoch and Jubilees and used at Qumran.¹¹ More recently, Colin Humphreys argued that Jesus and his disciples used a pre-exilic lunar calendar which, he claims, was still used by Galileans.¹² All attempts at harmonising the two traditions, however, are beset by the same four problems:

• • First, there is little evidence for any widespread use of alternative calendars in first-century Palestine. In a thorough overview of the much-fêted Qumran calendar, Jonathan Ben-Dov and Stéphane Saulnier show that recent scholarship has concluded that the calendar was neither strictly lunar nor solar, but rather composed of a rigid 364-day year.¹³ The fact that this calendar made no provision for intercalations is problematic, and, as Sasha Stern points out, prolonged use over an extended period of time would have quite substantially severed the link between agricultural festivals and the cycle of the crops. In all probability, he suggests, schematic calendars such as that at Qumran (and also those of 1 En. 72–82 and Jubilees) served not as living calendars, but as idealistic or theoretical models related to a future world order. In Stern's view, calendrical sectarianism had ceased by the first century.¹⁴

• • Second, Jesus' teaching shows no interest in the calendar, and both Jesus and his early followers in Acts appear to have visited the Temple at exactly the same time as other Jews. If they routinely used a different calendar, it is strange that it does not show up elsewhere in the tradition.¹⁵

• • Third, if Jesus had celebrated the Passover according to a different calendar, it seems very odd that one of his followers (usually thought to be John) blatantly dated the last events in his master's life not by his preferred system of reckoning, but by that used by his chief priestly opponents.

• • Finally, one further difficulty with the theories of Jaubert and Humphreys in particular is that by their reckoning the Last Supper was celebrated on either a Tuesday (so Jaubert) or a Wednesday (so Humphreys). Yet we have already seen that one of the few things that the Synoptic and the Johannine traditions agree on is that Jesus' last meal took place on a Thursday. By attempting to solve one discrepancy, these reconstructions have created another.

More plausible, perhaps, is the suggestion that the discrepancy lies in a difference in dating between Palestine and the Diaspora. Since the Passover was calculated following the sighting of the new moon (as we shall see below), and since there does not seem to have been a centralised body to endorse one particular calendar at this point (at least as far as we know), diversity could have existed among various Jewish groups.¹⁶ It is possible that Jews in Palestine observed Passover on Saturday that particular year, and that those in certain areas of the Diaspora may have celebrated it on Friday. In view of this, M. H. Shepherd suggested that John followed the testimony of Christians in touch with priestly circles in Judaea, while Mark followed traditions endorsed by his own Roman church.¹⁷ The difficulty with such a hypothesis, however, is that we would imagine that anyone making the journey to Jerusalem would automatically follow the Judaean calendar, celebrating the feast with the rest of the city. All traditions and recollections of events would simply follow the Jerusalemite dating. It is difficult to see why they would be transposed onto a calendar used by Roman Jews, particularly if the differences were simply to do with observation rather than sectarian debates.

Faced with these significant difficulties, most scholars accept that we simply have to choose between the two options. But which one—John or the Synoptics? Somewhat surprisingly, perhaps, recent scholars have tended to give preference to John.¹⁸ At first sight, John's scheme has two great advantages over that of the Synoptics: first, John's account is internally consistent while Mark's (as we shall see in a moment) is riddled with difficulties. Second, John's low-key account of Jesus' informal Jewish hearing on the day of Preparation chimes much more harmoniously with what we know of Jewish jurisprudence in the first century than the Synoptic record of a grand meeting of a Sanhedrin convening a capital case on the very night of Passover.¹⁹

Thus, by a significant modern consensus, the date of Jesus' death was 14th Nisan, and the task of the historian is to work out on what years the 14th Nisan fell on a Friday and then to convert it to the Julian calendar to give a date which is meaningful to us. The timescale can be narrowed quite considerably: if we accept the historicity of Luke 3.1 (which dates the beginning of John the Baptist's ministry to 28 at the earliest) and balance against this the need to fit in Pauline chronology, Jesus' death must have taken place some time between 29 and 34 CE.²⁰ The question then becomes more straightforward: are there any years between these margins when the day of Preparation fell on a Friday?

There are two ways of solving this puzzle: while an earlier generation of scholars relied on ancient calendrical tables,²¹ modern scholars have the benefit of computers. Whichever method is employed, however, astronomers can pinpoint the beginning of the months in the first century by locating the lunar conjunction (when the moon is between the earth and the sun) and then, by adding two weeks, find a date for the Passover full moon.²² There are of course a number of complications. The beginning of months in antiquity was calculated by observation of the new crescent moon in the evening sky; without specific climatic and atmospheric information, it is impossible to know exactly what would have been visible in Palestine.²³ Furthermore, it is not impossible that occasionally political or religious factors may have played a part in dating (for example the desire to avoid the day of Preparation of the Passover falling on a Sabbath). And finally, it is also clear that a lunar year of approximately 354 days needed an extra thirteenth month every few years in order to synchronize with the solar year; intercalations, however, seem to have been at the discretion of the Jewish high priestly rulers and so are difficult to factor into the figures.²⁴

The most recent calculations (those by Colin Humphreys and D. G. Waddington in the science journal Nature) claim to have surmounted all these difficulties, though their dates are almost entirely in line with earlier proposals.²⁵ Thus, 14th Nisan appears to have fallen on two possible dates within our broader time period: Friday 7 April 30 CE or 3 April 33. (By way of comparison, using the Synoptic dating, Nisan 15 fell on a Friday only in 27 CE and 34 CE, both of which are at the very margins of possibility.) It looks, then, as though astronomy has reinforced the Johannine tradition. Hence we have the two most popular dates for Jesus' death today, with those who favour a shorter ministry (or who date Luke 3.1–2 reasonably early) favouring 7 April 30, and the smaller group who favour a longer ministry opting for 3 April 33.

2. The Implications of this Date and the Evidence of Mark

At the heart of the most common dating of Jesus' death, then, is a preference for Johannine chronology, in which Jesus dies at the very moment that the paschal lambs were sacrificed in the Temple.²⁶ The crucifixion fits into a pattern whereby Jesus is shown from the very beginning of the gospel to be the replacement of Jewish feasts and institutions (the Temple, purification processes, the manna in the wilderness and so on). It is no great surprise to the reader to learn that now, at the feast of Passover, Jesus will take the place of the sacrificial lamb. As noted already, John's narrative is both internally coherent and consistent in its presentation of the timing of Jesus' death.

Yet I have to confess to being more than a little wary of placing too much historical reliability on John's account. The problem is not simply the theological use to which John puts his material: there is of course no reason why something should not be both theological and historical; it is perfectly possible for a good theologian to craft meaning from historical events. The difficulty is a methodological one. In a gospel not generally noted for its historical accuracy, it is unclear to me on what grounds we should prefer John's account at this point. It is undoubtedly the case that certain details here and there in the Fourth Gospel may well go back to historical reminiscences,²⁷ but I wonder whether the date of Jesus' death should really be assigned to this category. Surely a date which derived from Mark, our earliest and generally least tidy of the gospels, would be preferable. And a closer look at Mark, I suggest, is revealing.

Although Mark casts the last supper as a Passover meal, it is widely acknowledged that nothing in his account requires such a date. There is no reference to the lamb, to the bitter herbs, or to the recitation of the Exodus story.²⁸ The Passover connection is made by only two passages, both of which are generally regarded as redactional: 14.1a, which exhibits a clear Markan proclivity for a double time reference;²⁹ and 14.12–16, which is undoubtedly based on the earlier story of the colt in Mark 11.1–11.³⁰ Once these two passages are removed, nothing in the Markan passion narrative links the last supper—and therefore Jesus' death itself—to the day of Passover. Furthermore, there are elements within the Markan account which sit awkwardly with the evangelist's dating:

2.1. The Note that the Chief Priests and Scribes Decided not to Arrest Jesus during the Feast Lest There Be a Tumult of the People (Mark 14.2)

This is an odd passage for a number of reasons. Quite obviously, Jesus' followers could have had no reliable knowledge of the plots of the Jewish authorities. The only specific information given here is the decision of the priestly leaders to arrest Jesus, which was almost certainly inferred from subsequent events. The date of the intended arrest, however, is interesting. Adela Collins puts stress on the phrase ἐν δóλῳ (by deceit) and suggests that what the chief priests agree to avoid is arresting Jesus openly during the feast in case of riots; their use of Judas achieves this aim, and thus for Mark there is no discrepancy between their intentions and ultimate activities.³¹ More commonly, though, scholars do see a tension here between the plan of the Jewish authorities (not to arrest Jesus during the feast) and the way events actually turn out (arrest on the day of Passover itself). Assuming he noticed the inconsistency, it is possible to argue that Mark allowed it to remain so as to highlight the theological point that events worked themselves out according to God's plan, not those of Jesus' enemies.³² Even so, the discrepancy is intriguing, and it is hardly surprising that a majority of scholars are still inclined to accept the findings of the form critics who argued that this passage formed part of Mark's source, a source which dated Jesus' death rather differently to the chronology which Mark himself wished to promote.³³

2.2. The Release of Barabbas (Mark 15.6–15)

This is not the place to discuss the historicity of the supposed amnesty or the likelihood of a Roman governor releasing a dangerous prisoner at the whim of the crowd.³⁴ But the release of Barabbas is present in both the Markan and Johannine traditions and clearly goes back to an early, pre-gospel narrative of some sort or other. The release is commonly explained as an act of goodwill on the part of Pilate; but, as J. P. Meier notes, the gesture would lose all sense if the prisoner were not allowed to celebrate the feast with his friends and family.³⁵ The release must have allowed Barabbas enough time for his purification (whether the full week, or a reduced ‘emergency’ measure³⁶) and makes most sense from the prefect's point of view as soon after his arrival in the city as possible. Thus, whatever we make of the details of the account, the Markan location of the Barabbas episode on the day of Passover itself is curious.

2.3. The Note that Simon of Cyrene Was ‘Coming in from the Country/Field’ (Mark 15.21)

However we translate ἀπ᾽ ἀγροῦ, Simon's actions are conceivably problematic for Mark's date. If Simon has just come in from the field, the implication seems to be that he has been working, something which was forbidden on the day of Passover (Exod 12.16; Lev 23.7–8). If he has just come in from the country (whether Cyrene, or more likely the surroundings of Jerusalem) he may well have walked more than was permitted on the Sabbath and holy days (Acts 1.12; m. Erub. 4.3).³⁷ It is true that we have little information as to how these rules were actually practised in first-century Jerusalem, but the passage as a whole seems, once again, to sit awkwardly with Mark's dating.

2.4. Jesus' Burial (Mark 15.42–46)

Joseph of Arimathaea's pious actions make perfect sense against a Jewish background in which burial tended to be swift and allowing corpses to remain overnight was considered a curse (Deut 21.22–23). But Mark has him visiting Pilate, buying a shroud, and associating (in some way or another) with a burial—all on the day of Passover!³⁸ We are reminded about the approaching Sabbath, but the Passover is completely ignored. The easiest explanation for all of this is that while Jesus may have died on the day before a Sabbath, that is a Friday, that day had nothing to do with the Passover.

2.5. The Impression of a Holy ‘Week’ in Mark Is Much Less Secure than We Often Imagine.

The transition between the early days, particularly Monday through to Wednesday, is not at all clear, and the evangelist's arrangement of material is often topical rather than chronological (as was the case earlier in the gospel).³⁹ Historically, the interval between Jesus' triumphal entry and his Last Supper may have been significantly longer or even significantly shorter than a week. At all events, it is clear, as Adela Collins notes, that Mark is not particularly interested in chronology at this point.⁴⁰ He wants to show that Jesus died on the day of Passover, but has little interest in (or perhaps knowledge of) the time frame of events during Jesus' stay in Jerusalem.

Now I am by no means the only person to have highlighted these chronological complications in Mark. A routine trawl of the major commentaries will find them all listed and dealt with in various degrees. The most common conclusion to be drawn from them, however, is that they are indirect confirmation of the Johannine dating. Both Gerd Theissen and J. P. Meier, for example, argue for a pre-Markan passion narrative which exhibited a different chronology from that now found in Mark, but both assume that this agreed with John in dating Jesus' execution to the day of Preparation.⁴¹ While I think it is quite clear that a pre-Markan chronology is still visible within the text, however, I see no indication that this chronology aligned with what we have in John. Our exegetical survey of Mark revealed the following points: First, Mark seems to have been aware of an earlier tradition which did not locate the execution during the feast (Mark 14.2, a note which would count against John's chronology too, as the Jewish authorities would hardly have divorced the day of Preparation from the rest of the feast, and in any case the risk of a riot was equally high—perhaps even higher—on this particular day). Second, the story of Barabbas, if connected to the execution of Jesus at all historically, may provide weak evidence for the trial of Jesus taking place some time before the festival. Third, nothing aside from two clearly redactional passages (Mark 14.1a and 14.12–16) link the execution of Jesus to the day of Passover, and the accounts of Simon and Jesus' burial reinforce the view that Jesus died on an ordinary day. Taking all of this into account, it is difficult not to agree with Matti Myllykoski that the earliest, oral stages of the passion tradition were not articulated in terms of exact dates, and that Jesus was crucified and buried before the Passover festival.⁴²

Once the date of Jesus' death is cut loose from the traditional alternatives (the day of Passover/the day of Preparation), it is in theory just as possible that it occurred after the feast as before. Yet, quite apart from the questionable Barabbas episode, two factors incline me towards Myllykoski's preference for a date prior to the Passover. First, for security reasons it would have been better for both Jewish and Roman authorities to dispose of the threat posed by Jesus prior to the celebrations. Second, all our earliest Christian texts associate Jesus' death with the Passover. Psychologically, it seems to me, the close connection between Jesus' death and the feast is more likely if he was arrested in the midst of festal preparations, as people contemplated the meaning and significance of the feast, rather than afterwards, when pilgrims had begun to think about home and had perhaps already begun to leave the city. Certainty on this issue is clearly impossible, but the evidence seems to push in the direction of Jesus having been executed before the festivities began.

If this historical reconstruction is accurate how do we account for the presentations of Jesus' death in Mark and John? The chronological frameworks of the two narratives now appear to differ not only from one another but also from actual historical events. I shall address this question in the next section, drawing on recent studies of memory, both individual and collective. I hope to show that both the Markan and the Johannine chronologies with which we are familiar are based on theological reflections derived from the memory that Jesus died at around the time of the Passover.

3. Theology and Memory

To talk of a death ‘at Passover’ in the first century, just as now, did not necessarily mean that a person had died on Nisan 15, any more than for a Christian to talk of the death of a loved one ‘at Christmas’ means that he or she died on 25 December. The link with the feast, though, lends a certain poignancy to events. In our own context, a death at Christmas has a certain resonance—a loved one taken away at a feast where families come together, a life cut short at a festival celebrating new birth and hope. The connection with the feast adds an extra dimension to the death.

But how likely is it that the actual date of Jesus' death would be forgotten and/or replaced by a more theologically resonant one? The human memory has been studied exhaustively in recent decades, and the overriding picture which emerges is one of fragility and subjectivity. On an individual level, we tend to fill in the blanks, to make sense of what we see or hear, and to allow later information to blend into and inform what we think we remember. Contrary to much popular opinion, intense personal memories (often referred to as ‘flashbulb’ memories) are no more accurate than others (though we are often quite confident that they are). Deterioration and change in memory begins within hours, as we relive experiences and struggle to make sense of them. Over time, we may retain the gist of what happened, but not the specific details; in other situations we may become increasingly confident regarding inaccurate details, simply because we have told the story so many times (in these cases what we are remembering is actually only the last time that we told the story). Time indications seem particularly prone to corruption: often we may remember an event, but not precisely when it happened, or even who we were with at the time. Strikingly, too, memories are always connected to our current experience, and questions of self-identity in the present effortlessly and unconsciously shape what we think we remember of the past.⁴³

Even more significantly, however, memories function in a group context, as social memory, which binds individuals together. As numerous studies of traumatic situations have shown, difficult memories of the past are often replaced by groups—often quite unintentionally—by fictitious memories designed to ease the burden of the past, or to project an understanding onto the present.⁴⁴ As the earliest Christians came together to remember Jesus and to ponder the significance of his death, it would hardly be surprising if the immense weight of the Passover festival began to shape their stories: to inspire them to talk of covenants, of sacrifice, of the plan of God from long ago, and to encourage hopes that the visions of the prophets and the promises of Jesus himself were about to be realised. The reading of texts such as Isaiah 53 and Psalms 22 (21 LXX), 69 (68 LXX) in the earliest Christian liturgy could only have contributed to the sense that Jesus' death belonged in a meaningful way to the great story of Israel. Older stories were recycled, merged with recent traditions, and new memories created. In some liturgical contexts, the Eucharist might have evoked the earlier Passover covenant; in others, Jesus represented God's suffering righteous one; while others combined the Passover imagery with the story of the scapegoat on the Day of Atonement, viewing Jesus' sacrificial death as a means of removing sin.⁴⁵ Stories about Jesus were indirectly stories about the earliest Christians themselves, their relationship to what was past, and what they hoped for in the future.⁴⁶ At a time when dates could not easily be checked, there would have been very little to constrain the emergence of different recollections of the day of Jesus' death. Most followers of Jesus would have been willing to embrace the chronology of their particular group, especially as the eyewitnesses themselves began to pass away.⁴⁷ And what Christian would not want to embrace a more theologically meaningful date? As Doron Mendels perceptively notes, ‘The memory in itself is what counted in antiquity, and not necessarily the reality behind it’.⁴⁸

Jewish historiography shows a marked interest in the correlation of dates, though as James Barr has shown, biblical chronology tends to be far from historical in purpose.⁴⁹ The Hebrew Bible, for example, gives more detailed dates at the beginning rather than at the end—the reader can count back from Solomon to creation, but has little sense of how biblical events map onto secular history; and in the NT, Matthew's genealogy, despite its three precise sets of 14, does not enable us to determine the year of Jesus' birth.⁵⁰ The point of it all is not to furnish us with historical dates, but to underscore the conviction that a divine plan lies behind human history. Precisely the same motive lay behind Judas Maccabaeus' decision to purify the sanctuary on the same day that it had been profaned by foreigners (25th Kislev; 2 Macc 10.5), and Josephus' note that the Second Temple fell on precisely the same date as the First Temple had fallen to the Babylonians (9th Av; War 6.269). After the fall of Jerusalem, several Church Fathers regarded the destruction of the Temple as God's judgment on Jews for the death of Jesus—even though 40 years separated the two events.⁵¹ Eusebius went so far as to claim that Jerusalem fell at Passover; while this ignores the clear testimony of Josephus (which the Caesarean Bishop obviously knew), it reinforces the sense of divine retribution for the death of Jesus (H.E. 3.5.5–6). Other Church Fathers tried to link the date of Jesus' death with the spring equinox, generally believed to have been the day on which the world was created.⁵² It is clear that religious memory has never confined itself to the straightjacket of precise historical remembrance (even if such a thing were possible), but has always freely made connections in pursuit of deeper, theological truths.

If we turn to our earliest accounts of Jesus' death, we can see this process in its earliest stages. Undoubtedly Jesus' death was given a theological interpretation, inspired by its festal setting, right from the beginning (perhaps in the minds of some faithful onlookers even as he hung on the cross). By the 50s, Paul was a witness to what was probably a widespread understanding of Jesus as ‘our paschal lamb’ (τὸ πάσχα ἡμῶν), though in a context which did not require the apostle to give any specific date for the crucifixion (1 Cor 5.7). Clearly all that was important for Paul was to set Jesus' death against the backdrop of God's saving acts in the past, and to appreciate its theological significance. As the first passion narratives began to be composed, however, this theological understanding had to become more concrete. The tradition known to John placed Jesus' death at the very moment that the lambs were sacrificed in the Temple, casting him as the new paschal lamb, whose death removed the sins of the world.⁵³ A different tradition linked Jesus' last meal to the Passover, so that the eucharistic commemoration of Jesus' death now took the place of the Passover meal, and became the symbol of the new covenant between God and his people. This is the interpretation found in Mark and enhanced in the longer version of Luke 22.14–20. Thus, both the Johannine and Markan traditions narratively represent Jesus' death as profoundly meaningful, but both are based in the end not on any historical reminiscence, but on collective theological and symbolic elaborations of the memory that Jesus died ‘around Passover’.

4. Conclusion

I began by looking at the most common date for Jesus' death—7 April 30 CE—and considered its emergence and implications. I then offered an alternative (and in my view more likely) reading of the passion narratives in which Jesus' death actually took place some time (perhaps up to a week) prior to the day of Passover. I see no good reason to doubt that Jesus was executed on a Friday; the only thing that I want to challenge is the precise link between these events and the Passover. If I am correct, this means that we can no longer use astronomy to date Jesus' death. It is no longer a question of looking for 14th or 15th Nisan, because Jesus' death could have happened at any date round about then—perhaps the 11th or the 10th or the 9th Nisan. We simply have no way of knowing. And while it would be wonderful to have one secure date in the wild mass of competing Jesus traditions (comparable perhaps to the Gallio inscription in chronologies of Paul), I suspect that one of the main reasons why normally hard-headed Jesus critics sign up so quickly to 7 April 30 CE is because they desperately want it to be true. In the end, all that the evidence allows us to claim is that Jesus died some time around the Passover, perhaps a few days before the feast, any time between 29 and 34 CE.