2 Recent Grammatical Discussions of the Historic Present and Their Relevance to John 5.2

Before discussing the most recent grammatical insights into the historic present, I will first briefly introduce the relevant passage in John 5.2, in which the author uses the present tense to introduce the topographic information he provides for readers in his description of the Pool of Bethzatha. Describing Jesus’ attendance at a pilgrimage festival in Jerusalem, the author begins his narration of this event in the imperfect and aorist tenses:

Then the author goes on to provide his readers with topographic information, setting the scene for the unfolding events, and he puts this information in the present tense:

The narrative then continues in the imperfect tense:

The topographic information about the colonnaded Pool of Bethzatha is thus given in the present tense amidst the otherwise consistent use of the past tense (whether the imperfect or the aorist).

However, it does not make sense to view the use of the present tense in the statement ‘there is (ἔστιν) in Jerusalem […] a pool’ as an instance of the historic present. Regarding the historic present, some grammarians are of the opinion that the verb ‘to be’ (εἰμί), which is used in the Gospel’s statement (5.2), is never used as a historic present. Indeed Wallace made a remark to this effect in his 1990 article, stating that εἰμί is never used in this way in the Gospel of John,Footnote ¹⁴ and he repeated this view in his 1996 grammar with regard to all New Testament writings.Footnote ¹⁵ In response to Wallace, Andreas Köstenberger, in his 2004 commentary on John, gives as a counter-example Jesus’ statement in John 10.18: ‘All who came (ἦλθον, aor ind act 3rd pl) before me are (εἰσὶν, pres ind act 3rd pl) thieves and bandits; but the sheep did not listen (ἤκουσαν, aor ind act 3rd pl) to them.’Footnote ¹⁶ This present tense is real, however, because what matters here is the lasting validity of the comparison that the author draws between Jesus and those who came before him. Prior to Wallace’s article, in a specialist study of this Greek verb, Martín S. Ruipérez argued the same with regard to all Greek writings. In the current millennium, Albert Rijksbaron has addressed the same issue. According to Ruipérez, the historic present does not occur with verbs that describe a state, so that verbs such as εἰμί (‘to be’) and κεῖμαι (‘to be laid, to lie’) never occur in the historic present: ‘le présent historique n’existe pas dans les verbes d’état. Des verbes comme εἰμί “être”, κεῖμαι “être couché”, n’apparaissent jamais au présent historique.’Footnote ¹⁷ And according to Rijksbaron, historic presents ‘are, in fact, confined to telic, or terminative, and momentaneous verbs (so-called “accomplishments” and “achievements”, respectively), and do not occur, then, with durative-stative verbs. This means that forms like ἐστί, κεῖται, μένει, ἔχει, νομίζει, εὕδει are never used as a historic present.’Footnote ¹⁸ Furthermore, Rijksbaron believes: ‘The historic present is only found with terminative (telic), not with stative (atelic) verbs […] Thus, the present indicative of verbs like βασιλεύω, εἰμί, ἔχω, οἶδα, ῥέω is never used as a historic present.’Footnote ¹⁹ Very recently, in The Cambridge Grammar of Classical Greek, Rijksbaron and his co-authors have offered a slightly more nuanced view, stating that the historic present ‘occurs nearly exclusively with telic verbs (thus verbs such as εἰμί be, κεῖμαι lie, are not normally used as historical presents)’.Footnote ²⁰

Other scholars keep open the possibility that εἰμί (‘to be’) may be used as a historic present. In an elaborate critique of Rijksbaron, Gerard Boter challenges ‘the view that in Greek the use of the HP [historic present] is confined to telic verbs (or, for that matter, telic sentences)’.Footnote ²¹ He argues that passages do exist in which the historic present is also used with verbs that describe a state, and he presents two passages in the tragedies where the present tense of the verb εἰμί is best understood as a historic present.Footnote ²²

In his criticism of the categorical view that the historic present does not occur with atelic, stative verbs, Boter is to some degree joined by Arjan Nijk, the author of the most recent monograph-length study of the historic present. According to Nijk, the historic present ‘is in fact used with atelic verb phrases’, although he adds that ‘this use is relatively uncommon’.Footnote ²³ In a written clarification, however, which the present author has been given permission to quote, Nijk indicates that this statement can be qualified, and that it is important to differentiate between the larger category of atelic verbs, which also includes atelic verbs that denote processes and activities without an inherent endpoint (such as ‘to walk’ and ‘to pursue’), and the sub-category of atelic verbs that denote states (such as ‘to be’ and ‘to be laid, to lie’). With regard to the latter, in his study Nijk argues that verbs such as εἰμί (‘to be’) depict a state that is 1) atelic, i.e., without an inherent endpoint; 2) durative, i.e., the designated event lasts longer than a moment; and also 3) non-dynamic, i.e., steady over time.Footnote ²⁴ In his written clarification, Nijk adds that whereas the use of the historic present with the larger category of atelic verbs is ‘relatively uncommon’, its use with the sub-category of atelic verbs that describe states such as ‘to be laid, to lie’ and a few other incidental cases is ‘very rare’, and its use with ‘to be’ is even ‘non-existent’.Footnote ²⁵

Despite their remaining differences with regard to the verb ‘to be’, in their personal correspondence with the present author, Boter and Nijk agree that in the case of John 5.2, ἔστιν is not a historic present, but rather a ‘normal’ present. They are joined in this opinion by James Clackson (Cambridge), with whom the present author was able to discuss the matter.Footnote ²⁶ These scholars all agree that in the case of John 5.2, the present tense ἔστιν in the statement ‘there is (ἔστιν) in Jerusalem […] a pool, […] which has five porticoes’ is simply setting the scene for the current reality at the time of writing, the background against which the events took place.

Moreover, the formula that the author uses – ‘there is in (ἔστιν δὲ ἐν)’ followed by i) a particular locality in the dative case (‘in Jerusalem’, ἐν τοῖς Ἱεροσολύμοις) and ii) an architectural or geographic/geomorphological subject in the nominative case – is a formula that has been used since Herodotus (5th cent. bce) and occurs almost exclusively in works by historians, geographers, travel authors, ethnographers and natural historians. In this formula, ἔστιν is emphatically placed at the beginning of the sentence. In such a case, the uses of the present tense are i) ‘veridical’ (in statements about particular truths), ii) ‘existential’ (in statements about the current existence of particular things), or iii) ‘potential’ (in statements about current possibilities).Footnote ²⁷ The emphatic use of ἔστιν in the application of the formula in John 5.2 is an ‘existential’ use of the present, so that the use of the verb in the historic present can be excluded. This existential use of ἔστι(ν) at the beginning of a sentence is also explicitly acknowledged in The Cambridge Grammar of Classical Greek, which states: ‘Alongside […] 3 sg. ἐστι he/she/it is [without accent], there is a […] form ἔστι [with accent], used at the start of the clause, in the “existential” use there is’ (p. 290 §24.34 n.1).Footnote ²⁸

In what follows, I will provide the first passages in Greek literature where this formula is attested – in Herodotus – and summarise all the other instances I found. As we will see, the formula is never used to express a use of the historic present and, as in John, indicates the existence of particular physical realities at the time of writing.

The first instance of this formula in Herodotus occurs in a long description of the successive Egyptian kings (pharaohs), when Herodotus switches his tenses to the present in his description of Proteus:

In this historical narrative about King Proteus, Herodotus switches to the present tense when he states that there is ‘now’ (νῦν), i.e., at the time Herodotus is writing his work, a precinct of Proteus in Memphis, and that ‘there is in’ (ἔστι δὲ ἐν) that area of land a temple of Aphrodite.

Secondly, in another passage in his Histories, Herodotus switches to the present tense when describing how Phoenice, Palestine and Cyprus fit into all the satrapies that King Darius I of Persia created (The Histories 3.88–97) before his death in 486 bce, in the century in which Herodotus (who probably died in the 420s bce) was writing:

In this historical narrative about the satrapies Darius established, Herodotus changes his past tenses into present tenses when he describes how, as he believes, this division into satrapies still exists in his own time or, if this division had changed, how the regions that his readers now (i.e., in Herodotus’ time) know under the names of Phoenice, Palestine, and Cyprus were part of the fifth province (in Darius’ time).

Thirdly, in a final example, Herodotus applies the same formula when he describes the pass that leads through Trachis into Hellas and relates, ‘There are in (ἔστι δὲ ἐν) this pass warm springs for bathing, called by the people of the country The Pots (τῇ ἐσόδῳ ταύτῇ θερμὰ λουτρά, τὰ Χύτρους καλέουσι [pres ind act 3rd pl] οἱ ἐπιχώριοι), and an altar of Heracles stands thereby’ (Herodotus, The Histories 7.176.3).

In all of these instances, the formula ‘there is in (ἔστι[ν] δὲ ἐν)’ + locality in the dative case + an architectural or geomorphological subject in the nominative case, with its emphatically placed ἔστιν, makes an ‘existential’ assertion. The same is true in Pseudo-Scylax of Caryanda’s (4th cent. bce) travel accounts, in the work of the Jewish author of 2 Maccabees (later 2nd or early 1st century bce) and, in the Augustan era, in the accounts of the geographer Strabo of Amaseia (c. 64 bce–after 21 ce).Footnote ³⁰ Then the formula is also taken up in the New Testament writings, in the passage we are currently discussing – John 5.2.

Subsequently, the formula is used by Pausanias, the Greek travel writer (fl. c. 150 ce), who in his Description of Greece employs it countless times as his standard way of referring to the existence in particular locations of architectural (and artistic) structures such as houses, buildings, sanctuaries, temples, precincts, statues, busts, altars, fountains, wells, pyramid-shaped stones, tombs, theatres, marketplaces and – particularly relevant as an analogy for John 5.2 – bathing establishments (λουτρά).Footnote ³¹ This formula can also be used to point to the current existence of a geographic location in a particular region or a specific space in a building in a particular city,Footnote ³² and also to the existence of geomorphological features such as springs, lakes, caves, hills, plains and rivers.Footnote ³³

To give an example of the architectural structures to which Pausanias refers, with regard to Megara in Attica he writes:

What is relevant for our study of John 5.2 is the fact that Pausanias refers to the stone in the present tense (‘there is […] a stone’) and immediately identifies this stone by describing it with the aid of a participle in the present tense, as ‘a stone exhibiting (παρεχόμενος; masc nom pres part mid-pass sg) the shape of a small pyramid’. The same happens in John 5.2, where the author identifies the pool by describing it with the aid of a participle in the present tense as ‘a pool having (ἔχουσα; fem nom pres part act sg) five porticoes’. The present participles in these instances identify the nouns that are described in the present tense and that do not exist without these qualities of having a pyramid shape or having five porticoes. This is relevant for John 5.2 because the author does not simply state that the pool exists, but that the pool with the five porticoes exists. The same transpires in another passage in Pausanias, where he uses the same formula, stating: ‘There is on the left of the gateway a building (ἔστι δὲ ἐν ἀριστερᾷ τῶν προπυλαίων οἴκημα) having (ἔχον; neut nom pres part act sg) pictures’ (1.22.6).

That the formula in question refers to an architectural structure which is described as existing at the time of writing is true regardless of the genre of the text. Although the formula is used almost exclusively by historians, geographers, travel authors, ethnographers and natural historians, it is also employed by the novelist Achilles Tatius (late 2nd cent. ce) to set the scene in his Leucippe and Clitophon. Mapping the world is generally important in Greek novels and is usually undertaken by means of travel accounts, geography and natural history.Footnote ³⁴ However, as Koen De Temmerman has noted, Tatius’ novel ‘is the oldest of the extant novels that emphatically draws attention to the importance of an elaborate representation of space’, which is ‘structured around three geographical areas’, including Egypt, all of which are ‘connected by sea voyages’.Footnote ³⁵ Moreover, as Tim Whitmarsh indicates, Tatius sets his work of fiction, which is at the same time so ‘compendious’ and ‘kaleidoscopic’, in the ‘real’ world.Footnote ³⁶ One example of this is Tatius providing the following setting for an episode in his narrative that takes place in Egypt, stating: ‘There is in Pelusium’ a particular statue of Zeus, which constitutes the backdrop for the actions of the two protagonists:

This passage starts off by sketching an existing situation, and the narrative is described against this background in past tenses. The existence of a temple of Zeus Casius in Pelusium is attested by Strabo and Sextus Empiricus,Footnote ³⁷ and the authenticity of the Zeus statue has been archaeologically confirmed.Footnote ³⁸ Thus this reference to the Zeus statue at the temple in Egyptian Pelusium is a reference in the present tense to an existing structure as the place where the action subsequently described in past tenses took place in the past. This is exactly what happens in the Gospel of John when the author refers in the present tense to the Pool of Bethzatha as an extant architectural setting where the action took place. The fact that Tatius’ Leucippe and Clitophon is a novel makes no difference here, because the point is that the author provides an extant historical setting for his narrative. After Tatius, the typical application of this formula to history and natural history is again found in Cassius Dio (c. 164–after 229 ce) and in Aelian (161/177–230/238 ce).Footnote ³⁹

Thus it appears that, without exception, all the main instances of the formula ‘there is in (ἔστι[ν] δὲ ἐν)’ + locality in the dative case + an architectural, geomorphological or natural historical subject in the nominative case in Greek literature since Herodotus, regardless of genre (whether (natural) history, geography, ethnography or novel), refer to an object that is in existence at the time of writing. In all of these cases the verb ἔστι[ν], emphatically placed at the beginning, is never an instance of the historic present, but always has an ‘existential’ meaning, in the sense that all of these instances refer to a currently extant object. These analogical applications of the formula strongly suggest that in the statement ‘there is in Jerusalem […] a pool (ἔστιν δὲ ἐν τοῖς Ἱεροσολύμοις … κολυμβήθρα) […], which has five porticoes’ (5.2), ἔστιν also has this kind of existential meaning and refers to a monumental colonnaded pool complex that currently exists and is (still) standing at the time of writing. It therefore seems that the onus of proof is now on those who would continue to argue that ἔστιν in John 5.2 is an instance of the historic present – they must justify why only this particular use of the formula, amongst all of its other occurrences in Greek literature, would constitute a historic present, and explain what function this unique use of ἔστιν would serve in the discursive and literary context of the Gospel. Based on all the analogies of the formula, however, the use of ἔστιν must be existential, and it functions to set the scene for a narration of historical events that, according to the author of the Gospel, occurred against this backdrop. This is not inconsistent with the fact that, in the rest of the narrative about the healing at the Pool of Bethzatha, the author does sometimes shift into the historic present, as happens often in this Gospel, especially when highlighting Jesus’ speech (5.6, 8) or describing how Jesus looks for the man after these events and finds him again (5.14). Apart from such exceptions, however, the narrative remains in the past tense. Even though the narrative in John 5 contains examples of the historic present, this does not necessarily mean that the use of the present tense ἔστιν in 5.2 is historic. In fact, if we consider all the other uses of this formula, then it is clearly not.

To circumvent the force of the existential meaning of ἔστιν in John 5.2, it might seem possible to draw a parallel with the fact that Josephus and some other authors, such as Clement, continue to refer to the Jerusalem temple in the present tense even after its destruction in 70 ce. However, as I will show, these references in the present tense, insofar as they do indeed occur in these passages, do not concern the temple buildings, but rather the temple cult, and constitute uses of the habitual present, indicating the expectation (still justified at the time of writing) that the Romans would sooner or later allow the Jerusalem temple to be rebuilt so that the temple cult could be resumed.

In both Josephus’ Jewish War and his Jewish Antiquities, all the descriptions he gives of the Jerusalem temple concern the temple buildings and are given in the imperfect, aorist and pluperfect tenses. In his Jewish War, Josephus describes Jerusalem (5.136–247). He makes no use of the present tense in his description of the temple buildings, but when he describes construction works in the city, he uses the historic present once: to describe the attempt made by Agrippa I (r. 41–44 ce) to build a third wall around the extramural Bezetha area of the city (5.147–54).Footnote ⁴⁰

The same is true for the descriptions of the temple buildings in Josephus’ Jewish Antiquities (15.380–425), which are given in imperfect, aorist, and pluperfect tenses. Again Josephus makes no use of the present tense in describing the temple buildings – except once, when he uses the historic present to ‘punctuate’ his narrative concerning particular, distinct phases in the process of building the temple since Solomon’s time (15.401).Footnote ⁴¹ As The Cambridge Grammar of Classical Greek indicates, sometimes the historic present ‘is not so much used at dramatic turns, but rather to “punctuate” a narrative, dividing it up into separate sections by highlighting each new step’ (p. 431, §33.55). A second instance of the present tense occurs when Josephus describes the high priest’s activities (15.403).Footnote ⁴²

The situation is different in his Against Apion, where Josephus focuses on the temple cult. In a section that has only been preserved in Latin, Josephus describes the temple buildings in the imperfect tense, but then moves to the predominant use of the present tense in describing the temple cult (2.102–9) and emphasising how ‘careful is the provision for all the details of the service’ (Tanta uero est circa omnia prouidentia pietatis; 2.105). John Barclay has also observed this, saying, ‘Without drawing attention to the matter, Josephus describes the temple structures in the past tense, but the laws governing access (2.105–6) and the cultic activity (2.108–9) in the present tense (cf. 2.76–7).’Footnote ⁴³ But whereas Barclay believes the reason for Josephus’ description of the temple buildings in the past tense is that this is ‘rhetorically appropriate in relation to events at the time of Antiochus’ (IV Epiphanes) that Josephus is describing,Footnote ⁴⁴ I believe the reason for the past tense is rather that Josephus is writing after the destruction of the Herodian temple buildings.

Similarly, in a later section of Against Apion that has been preserved in the original Greek, Josephus continues to use the present tense in his description of the temple cult when he says, ‘The priests are continually engaged in His worship (τοῦτον θεραπεύουσιν μὲν διὰ παντὸς οἱ ἱερεῖς), under the leadership of him who for the time is head of the line (ἡγεῖται δὲ τούτων ὁ πρῶτος ἀεὶ κατὰ γένος)’ (2.193), and ‘Our sacrifices are not occasions for drunken self-indulgence … but for sobriety (θύομεν τὰς θυσίας οὐκ εἰς μέθην ἑαυτοῖς …, ἀλλ᾿ εἰς σωφροσύνην)’ (2.195). However, he also uses the future tense here, stating, ‘With his colleagues he will sacrifice to God, safeguard the laws, adjudicate in cases of dispute, punish those convicted of crime’ (2.194).Footnote ⁴⁵ Clearly, the reason for the use of the future tense in 2.194 is that the temple is still a ruin at the time of writing, and so Josephus, anticipating its reconstruction, describes the resumption of the temple cult in the future tense. When Josephus does use the present tense with regard to the Jerusalem temple in 2.193 and 2.195, it is important to note that such occurrences, like the instances of the future tense in 2.194, concern the temple cult, and that the present tense is never used to describe the temple buildings. Therefore, this use of the present tense to describe the temple cult does not provide an analogy for the use of the present tense in John 5.2, which does concern a building. Josephus’ use of the present tense to describe the temple cult is ‘habitual’, describing what has always been and will again be the case.

This present tense does not describe an ‘ideal’ situation of some ‘ideal’ temple cult, but is rather a habitual use of the present that describes the normal temple cult’s habitual situation.Footnote ⁴⁶ It is important to emphasise that after the Jerusalem temple was destroyed, Jews would indeed have expected that the Romans would give permission to rebuild it in due course, so that the temple cult would be resumed. At the time Josephus was writing, the Romans had diverted the temple tax that the Jews were paying to support their temple for the purpose of rebuilding the Romans’ own temple of Jupiter on the Capitol in Rome,Footnote ⁴⁷ which had just been destroyed in 69 ce, during the civil war between the competing candidates who sought to replace Nero after his death in 68 ce.Footnote ⁴⁸ The Jewish expectation would have been that at some future time, once the Capitoline temple of Jupiter was fully financed and rebuilt, the Romans would allow the Jewish temple to be rebuilt.Footnote ⁴⁹ Therefore, Josephus had every reason to use the habitual present tense in describing the Jerusalem temple cult, as this habit was expected to resume and continue in the future, in a newly built temple.Footnote ⁵⁰ Hopes for such a reconstruction were only quashed when Emperor Hadrian built a new temple on the Temple Mount in Jerusalem – not for the Jewish god, but for Jupiter,Footnote ⁵¹ although even the Mishnah still preserves the expectation that the temple will be rebuilt.Footnote ⁵² This also explains the use of present tenses in the description of the temple cult in the Letter to the Hebrews (5.1–4; 7.28; 8.3; 9.6–7, 25; 10.11; 13.10–11) and in the First Letter of Clement to the Corinthians (40.1–41.4), written in Rome in the mid-90s ce.

Hence, the use of the present tense in descriptions of the Jerusalem temple cult in Josephus, Hebrews and 1 Clement is habitual, a qualification which does not apply to the present tense used in the statement in John 5.2 that there ‘is’ a colonnaded pool in Jerusalem; the latter use of the present tense is existential, not habitual, because buildings cannot temporarily interrupt their existence, whereas habits can be interrupted and resumed.Footnote ⁵³ The present tense in John 5.2 is thus neither historic nor habitual, but is a present tense that describes the existence of a particular building at the time of writing. This existential sense is the sense in which the verb occurs in all the examples we have seen of the formula ‘there is in (ἔστι[ν] δὲ ἐν)’ + locality (in the dative case) + an architectural structure as subject (in the nominative case).

At the risk of being accused of making too much of the single occurrence of the verb ἔστιν in John 5.2, I would note that this is how one goes about reconstructing the date when a text was written on the basis of its internal evidence. Indeed Friedrich Blass, the author of a well-known New Testament Greek grammar and professor of classical philology at the University of Halle, recognised this when he drew an analogy between the ἔστιν in John 5.2 and the ἔστιν in an anonymous Hellenica that had recently been found at Oxyrhynchus in Egypt (in 1906), the so-called Hellenica Oxyrhynchia (P. Oxy. 842).Footnote ⁵⁴ The use of ἔστιν in its description of the Persian king was and still is understood ‘to provide a key indication that the text was written while the Persian Empire was still in existence, that is, before about 333 bc’.Footnote ⁵⁵ As Blass himself argued, ‘What is valid here ought also to be equally valid for John.’Footnote ⁵⁶ And just as the Persian Empire ceased to exist after 333 bce, so in all likelihood the colonnaded Pool of Bethzatha ceased to exist after 70 ce.

3 The Likely Destruction of the Colonnaded Pool in the Double Destruction of the Bezetha Area by Cestius (November 66 ce) and Titus (May 70 ce), and in the Stripping of the Area for Timber for the Roman Earthworks (June–September 70 ce)

We have now seen that ἔστιν in John 5.2 is a real, existential present tense and refers to the existence of the colonnaded Pool of Bethzatha at the moment of writing. In this and the following section, we shall consider the relevance of this present tense for the dating of the Gospel. Until what point in time did the colonnaded pool – the pool with the five porticoes – exist? We know from Eusebius of Caesarea (c. 260–339 ce) that by 314 ce, the Pool of Bethzatha had lost its five porticoes. Eusebius, who was a long-term resident of Caesarea Maritima, was personally acquainted with the local situation in Jerusalem and visited Jerusalem, for instance, to consult the church’s library archives while writing his Ecclesiastical History.Footnote ⁵⁷ In his Onomasticon, dated to 314 ce,Footnote ⁵⁸ Eusebius describes ‘Bezatha’ as follows:

Thus, Eusebius specifically reveals that the five porticoes mentioned in the Gospel indeed no longer existed by 314 ce. As Eusebius is writing before the Christianisation of Jerusalem (324 ce) under Constantine I (r. 306–337 ce, alone from 324 ce),Footnote ⁵⁹ he is effectively describing the situation that obtained after Hadrian (r. 117–138 ce) refounded Jerusalem as Aelia Capitolina in 130 ceFootnote ⁶⁰ and before the Byzantine Church of St Mary ‘in Probatica’ was built over part of the twin pools in the mid-fifth century ce.Footnote ⁶¹ Whereas in Eusebius’ time the pool’s five porticoes have disappeared, so that the former pool complex can indeed only be pointed out ‘in the twin basins’ (ἐν ταῖς […] λίμναις διδύμοις), the author of the Gospel, familiar with the Pool of Bethzatha when it existed prior to 70 ce, talks about the pool with its five porticoes in the singular: for him it is a single complex. Strictly speaking, as Gibson and Gurevich have shown, only the southern basin was used as a (purification) pool, whereas the northern basin merely functioned as a reservoir which collected the run-off rainwater and from which the immersion pool was fed.Footnote ⁶² Once the porticoes that gave the pool complex its unified appearance had gone, this pool could indeed only be attested in its two basins. These basins continued to exist, as they were hewn into the bedrock of the Bezetha valley,Footnote ⁶³ but the porticoes around them did not.

The situation Eusebius describes is also confirmed by another local author and eyewitness, Cyril of Jerusalem (bishop of Jerusalem, 348–86 ce), who refers to the colonnaded pool in the imperfect tense,Footnote ⁶⁴ and also by a fragment that has been attributed to either Theodore of Mopsuestia (c. 352–428 ce) or even to Origen (c. 185–c. 255 ce), which also references the porticoes in the imperfect tense.Footnote ⁶⁵ Therefore Gibson, the author of the preliminary report on the excavations at the Pool of Bethzatha, is mistaken in his view that, ‘Judging by the testimony of Origen, the original four porticoes running around the edges of the twin pools with another across the middle, were still visible to visitors in his day (ca 231). This information was repeated by Cyril of Jerusalem before 348, but the language of Eusebius suggests that in his day, before 331, the actual porticoes were already in ruins.’Footnote ⁶⁶ Instead, the collective evidence from Eusebius, Cyril of Jerusalem and Theodore of Mopsuestia/Origen reveals that by 314 ce (or even by the mid-third century ce, if Origen is involved), the five porticoes had disappeared.

However, there seems to be no particular motive for removing the porticoes between 70 and 314 ce if they were still standing after the end of the First Jewish Revolt and the destruction of Jerusalem in 70 ce. If the colonnaded Pool of Bethzatha had survived the First Jewish Revolt, then no major historical events could explain why its porticoes disappeared between 70 ce and the time of the post-Hadrian Jerusalem described by Eusebius in 314 ce. Firstly, after 70 ce, the Romans had no reason to remove the porticoes if they were still standing. Between 70 and 130 ce, Jerusalem was the legionary base of the Legio X Fretensis, which would have enjoyed a colonnaded pool. The Jewish Diaspora Revolt of 116–117 ce did not affect Jerusalem, and the city was not attacked. Secondly, between 130 ce and the time of Eusebius, Hadrian and his successors would have had no reason to remove the Herodian Greek-style porticoes of the pool and thus leave it uncolonnaded. They probably also used the pool for their Asclepian healing practices, as indicated by ex-voto dedications found nearby,Footnote ⁶⁷ which have been dated to the second half of the second or to the third century ce.Footnote ⁶⁸ The Second Jewish Revolt of 132–136 ce, in which Bar Kokhba revolted against Rome, did not reach Jerusalem and thus cannot be the reason for the destruction of the porticoes either.Footnote ⁶⁹ Thirdly, Eusebius’ evidence for the disappearance of the porticoes dates from 314 ce, before Constantine’s first act in the Christianisation of Jerusalem – the construction of the Church of the Holy Sepulchre, which was consecrated in 335 ce. But if the porticoes were still standing at this time, Constantine would not have destroyed them either, because for him they would have been part of the pool complex at which, according to the Gospel of John, one of Jesus’ miracles had taken place.

As there are no particular reasons or motives for the Romans to have removed the porticoes between 70 and 314 ce, provided they were still standing in 70 ce, it seems that the more plausible explanation for their disappearance lies in the tumultuous events of the siege and gradual destruction of Jerusalem during the First Jewish Revolt (66–70 ce). The area was destroyed on two occasions, and subsequently, the Romans searched the broader environs for timber for the last phase of their siege. The first destruction of the Bezetha area, where the Pool of Bethzatha was standing, took place soon after the First Jewish Revolt broke out in the spring and summer of 66 ce. In October and November of that year, the governor of Syria, Cestius Gallus, intervened in Judaea, fighting his way up to Jerusalem, driving the revolutionaries away from the Jerusalem suburbs and into the city, and moving into the Bezetha neighbourhood. Agrippa I had attempted to enclose this area to the north of the Temple Mount with a third wall in 41–43/44 ce (Josephus, BJ 5.147–52, cf. 2.218–19; Ant. 19.326–7), but work on the wall had been suspended due to Emperor Claudius’ disapproval (BJ 5.152; Ant. 19.326–7). In November 66 ce, Cestius laid waste to this part of the city, which had in fact remained outside the first and second city walls:

Given this destruction of the Bezetha area in November 66 ce, it is questionable whether the colonnaded Pool of Bethzatha escaped the same fate. The effect of setting fire to porticoes is disastrous. Analogies show that it is relatively easy to destroy porticoes by setting the woodwork of their roofs on fire, for instance, by filling ‘the space between the rafters of the […] portico and the ceiling beneath them with dry tinder, along with bitumen and pitch’ (BJ 6.178), so that the roofs collapse and then dislocate the columns.Footnote ⁷⁰ Since Josephus describes how Cestius destroyed the Bezetha area outside Jerusalem’s city walls by fire at the very beginning of the Jewish Revolt in the autumn of 66 ce, it seems not improbable that the Pool of Bethzatha with its five porticoes also fell victim to this destruction.

Nevertheless, if the colonnaded pool survived the initial destruction of the area in 66 ce, the chances that it survived the First Jewish Revolt are further reduced by the fact that this neighbourhood was destroyed a second time in May 70 ce. By then, having most likely started work after Cestius’ withdrawal in 66 ce, the Jews had finished the third wall that Agrippa I (BJ 5.155) had begun building. In May 70 ce, Titus made an assault on the third wall (BJ 5.258–260), which from the Roman outsiders’ perspective was the ‘first’ wall. Once they had taken this wall, they destroyed the Bezetha area for a second time, as Josephus explicitly remarks:

And still, the destructions of this part of the city were not yet over; as Josephus repeatedly indicates, while preparing their further assaults, the Romans lacked timber with which to construct their earthworks, which they would throw up against the walls of the inner city and the Antonia fortress on the north-western corner of the Temple Mount, overlooking the Bezetha area Titus had already given.Footnote ⁷¹ During their assault on the third (or first) wall surrounding the Bezetha area,

The second wall was quickly taken after the first (BJ 5.331–47), but for the earthworks they constructed between approximately 30 May and 16 June 70 ce for their assault on the first (or, from the Romans’ perspective, the third) wall of the upper city and the Antonia fortress (BJ 5.347, 356–7, 466–7), the Romans needed much more material and timber. Discouraged by the Jewish counterattacks that destroyed their earthworks (BJ 5.469–70, 5.473), according to Josephus, Titus ‘pointed out the extreme difficulty of throwing up earthworks owing to lack of timber’ (βάλλεσθαι δὲ χώματα δύσεργον ἀπέφαινεν ὕλης ἀπορίᾳ, BJ 5.496) and decided to circumvallate the entire city first (BJ 5.499, 504); this circumvallation wall also ran through the Bezetha area: ‘he directed the wall towards the lower region of the New City (ἐπὶ τὴν κατωτέρω Καινόπολιν ἦγε τὸ τεῖχος)’ (5.504). Subsequently, the construction of earthworks resumed, although they still faced the challenge of a lack of timber:

This means that the Roman troops were searching for timber within a radius of ‘ninety stades’ (i.e., approximately 16 kilometres or 10 miles) from the Antonia fortress, which was situated opposite the Bezetha area.Footnote ⁷³ As the roofs of Greco-Roman porticoes consisted of timber and wood,Footnote ⁷⁴ it seems very likely that the Romans would have decided to use the timber from the five porticoes of the nearby Pool of Bethzatha, which had either already been affected by the fire Cestius started in the neighbourhood in 66 ce or destroyed by Titus in May 70 ce, after he had conquered the walls around the Bezetha area, or were only now dismantled for their timber. During their assaults, the Romans made use of local timber by felling trees and by destroying existing structures. For instance, when Sulla attacked Athens in 87 bce, he not only cut down the trees in the sacred grove of Plato’s Academy in the extramural suburbs of Athens, from which he built his siege engines, but also ‘knocked down the Long Walls, and used the stones, timber, and earth for the construction of the siege mound (λίθους καὶ ξύλα καὶ γῆν ἐς τὸ χῶμα μεταβάλλων)’ (Appian, Roman History 12; The Mithridatic War 121).Footnote ⁷⁵ Given the serious lack of timber available for the siege of Jerusalem, it seems implausible to assume that the porticoes of the Pool of Bethzatha would have remained intact. The pool was at a distance of only about 200 metres from the earthworks erected against the Antonia fortress,Footnote ⁷⁶ and the Romans were searching for timber in a broad radius of approximately 16 kilometres (10 miles). The pools around and between which the porticoes had been built measured 40 x 53 metres (the northern pool) and 47 x 52 metres (the southern pool),Footnote ⁷⁷ and the porticoes were probably 2.5 metres wide (if the analogy with the Pool of Siloam is valid),Footnote ⁷⁸ which would offer the Romans significant amounts of timber. It seems very unlikely that the porticoes of this colonnaded pool in the vicinity of the Antonia fortress would have remained intact through both of the above-mentioned destructions and also this urgent search for timber. According to Josephus, the suburbs and the surrounding countryside were reduced to a wasteland: ‘sites formerly beautified with trees and parks now reduced to an utter desert and stripped bare of timber’, the ‘beautiful suburbs’ desolated, as ‘the war had ruined all the marks of beauty’ (BJ 6.6–8).

This lack of timber continues for the rest of the siege throughout the summer of 70 ce. According to Josephus,

After the Roman army had overthrown the foundations of the Antonia fortress and raised new earthworks against the first wall, the work proved very demanding; they began collecting timber from an expanded radius of ‘a hundred stades’ (i.e., approximately 18.5 kilometres or 11.5 miles) (BJ 6.149–52). Even in early September 70 ce, when the Romans were preparing their attack on the upper city, Titus,

Given these descriptions of the Romans’ assaults on Jerusalem in the period from May to September 70 ce and their continuous lack of timber, which they procured from an ever-increasing radius of 16–18.5 kilometres (10–11.5 miles), it seems improbable that the timber-rich porticoes of the Pool of Bethzatha in the vicinity of the Temple Mount would have remained intact and untouched amidst the siege and capture of Jerusalem, provided they had not already been affected by Cestius’ fire or Titus’ destruction of the Bezetha area prior to the siege.

Moreover, what would have remained of Jerusalem after the final conquest and burning of the city at the end of September 70 ce (BJ 6.407–8) was destroyed when Titus ‘ordered the whole city and the temple to be razed to the ground’ (κελεύει … ἤδη τήν τε πόλιν ἅπασαν καὶ τὸν νεὼν κατασκάπτειν), with the exception of the three famous towers of Herod’s palace on the western wall, along with the part of the wall that was used to protect the camp of the Legion X Fretensis, which stayed behind in Jerusalem (BJ 7.1–4).Footnote ⁷⁹ ‘All the rest of the wall encompassing the city was … completely levelled to the ground’ (τὸν δ’ ἄλλον ἅπαντα τῆς πόλεως περίβολον … ἐξωμάλισαν οἱ κατασκάπτοντες, 7.3).

This also opens up another unexpected perspective that confirms the pre-70 ce dating of John 5.2. In the full passage, the Gospel’s author not only describes the Pool of Bethzatha in detail as a colonnaded pool with five porticoes, but also states that it is located ‘near the Sheep [Gate]’ (ἐπὶ τῇ προβατικῇ): ‘There is in Jerusalem near the Sheep [Gate] a pool, which is called Bethzatha in Hebrew, with five porticoes’ (ἔστιν δὲ ἐν τοῖς Ἱεροσολύμοις ἐπὶ τῇ προβατικῇ κολυμβήθρα ἡ ἐπιλεγομένη Ἑβραϊστὶ Βηθζαθά, πέντε στοὰς ἔχουσα). If we accept this reading of the Greek text in Neste-Aland’s twenty-eighth edition, then it is likely that the localisation ἐπὶ τῇ προβατικῇ refers to the Sheep Gate, as the phrase seems to be translated in virtually all modern bible translations by adding the missing female noun πύλη (‘gate’): ἐπὶ τῇ προβατικῇ πύλῃ.Footnote ⁸⁰ Gates in the city walls were indeed great topographical markers in antiquity and functioned as the equivalent of modern postcode areas. In order to localise temples, shrines, gymnasiums, houses, theatres, race courses, tombs, monuments and herms, authors such as Strabo and Pausanias inform their readership about the proximity of such buildings, facilities and objects to the gates where one enters the city in question.Footnote ⁸¹ These descriptions of buildings near particular gates also sometimes make use of the ‘there is in …’ formula employed in John 5.2 and may provide further details concerning these buildings, such as that they are surrounded by porticoes – as is the case in John 5.2.Footnote ⁸² Given this topographical practice of specifying buildings by describing their appearance and pinpointing their location by giving their proximity to particular city gates, it seems very likely that ‘Gate’ is the natural way of supplementing the phrase ἐπὶ τῇ προβατικῇ in John 5.2. The Sheep Gate (ἡ πύλη ἡ προβατική) is mentioned in Nehemiah as the gate that was built in this part of Jerusalem, on the north-eastern side of the Jerusalem temple, as part of the rebuilding of the walls which Nehemiah instigated in the Persian period (Nehemiah 3.1, 32 and 12.39; LXX 13.1, 32; 22.39).Footnote ⁸³ This gate, which lay in the small valley between the Temple Mount and the baris (the Hasmonean fortress immediately to the north of the Jerusalem temple), seems to have disappeared in the time of Herod the Great, in the 20s bce, when Herod extended the Temple Mount and turned the baris into the Antonia fortress, which was immediately connected to the new Herodian temple walls.Footnote ⁸⁴ According to Dominique Marie Cabaret’s recent suggestion, to compensate for the loss of the eastern gate into the city, Herod inserted a new gate in the second wall, to the west of the Antonia: the so-called Ecce Homo gate at the present Via Dolorosa.Footnote ⁸⁵ However, when Agrippa I began to build the third wall around the Bezetha area between 41 and 43/44 ce, it also needed gates on the eastern side, to enable traffic to leave the new town to the east as well,Footnote ⁸⁶ towards the Mount of Olives and Bethany, and – further afield – towards Jericho and the Decapolis.Footnote ⁸⁷ Although the construction of the third wall could not be completed due to the Romans’ disapproval, it probably had nine gates; two of these were placed in the eastern part of the wall, with one in a direct line from the Ecce Homo gate,Footnote ⁸⁸ so that travellers arriving from the east and entering Jerusalem through this gate would be led between the northern Temple Mount wall and the Antonia on their left-hand side, and the colonnaded Pool of Bethzatha on their right-hand side. This gate was very close to the location of Nehemiah’s original Sheep Gate, and so one can imagine that this name was transferred to the new gate in the third wall. If this is correct, then it also implies that the author, by localising the colonnaded Herodian pool for his readers in the direct vicinity of the Sheep Gate in the third wall, was writing after the wall’s erection, i.e., after 41–43/44 ce. The entire third wall, along with its gates, was subsequently destroyed by the Romans in 70 ce, remained in ruins between 70 and 130 ce, and was then hastily rebuilt due to (the threat of) the Second Jewish Revolt against Rome in 132–6 ce.Footnote ⁸⁹ Thus the Gospel’s description of the Pool of Bethzatha as located ‘near the Sheep [Gate]’ is only possible either before 70 ce or after 130 ce, but not between 70 and 130 ce. As we have seen, however, the post-130 ce pool of the Hadrianic era, as described by Eusebius, had retained only its two basins and lost its five porticoes. The full, combined description of the Pool of Bethzatha in the present tense as a pool that has five porticoes and is located near the Sheep Gate (in the third wall) is therefore only possible prior to 70 ce; thus we can date the Gospel with some degree of probability – bordering on certainty – to the pre-70 ce period.

Josephus’ reports on the double destruction of the Bezetha area by Cestius and Titus, the subsequent plundering of the entire area for timber and material for the earthworks that the Romans threw up to capture the city and the final destruction of the city are entirely consistent with archaeological findings in the Bezetha or New City (BJ 2.530, 5.504) area.Footnote ⁹⁰ Excavations undertaken by Yuval Baruch and Gideon Avni in part of the New City area led them to the conclusion that this area appears to have been abandoned after 70 ce.Footnote ⁹¹ Similarly Hillel Geva, who offers an overview of the findings based on the archaeological evidence from the New City by focussing on the pottery, has confirmed that ‘the development of this new area of Jerusalem came to a quick end’ in 70 ce.Footnote ⁹²

This all seems to suggest that the porticoes at the Pool of Bethzatha were destroyed in the course of 66–70 ce: either burned down by Cestius when he set fire to the Bezetha neighbourhood in November 66 ce, destroyed by Titus when he captured the third wall around the Bezetha neighbourhood and razed this area in May 70 ce, dismantled by the Roman troops in urgent need of timber for the construction of their earthworks in May–September 70 ce, or destroyed in the final post-victory destruction of Jerusalem. It is therefore wrong to suggest, as for instance, J. Ramsey Michaels does in his 2010 commentary on John, that the Gospel’s author uses the present tense in 5.2 because the pool still exists after 70 ce.Footnote ⁹³ John 5.2 is not about the pool, but about the colonnaded pool, the pool with five porticoes; in the absence of reasons why these porticoes would have been destroyed between 70 and 314 ce, their disappearance is most plausibly explained by the events of 66–70 ce, which provide us with reasons and motives for their destruction and/or dismantlement.

If the porticoes did survive and remained standing after 70 ce, against all odds, then one would expect that, in the face of the sheer extent of Jerusalem’s destruction, the author of the Gospel of John would at least have specified (from a post-70 perspective) that despite everything, the colonnaded pool was ‘still’ standing. The author could easily have used a formula like the one geographers and historians such as Herodotus, Strabo, Pausanias and Diodorus Siculus employed when they developed a particular historical perspective by describing a long lapse of time and/or many changes or even destructions that had taken place, despite which particular places, buildings or objects survived. Such formulas could have included: ‘there is also now still’ (ἔστι δὲ καὶ νῦν ἔτι),Footnote ⁹⁴ or: ‘up to now’ (μέχρι νῦν).Footnote ⁹⁵

Instead of using such formulas, however, the author of the Gospel chose the simple formula that indicates the current existence of a colonnaded pool, with no hint of a complicated situation despite which the pool ‘still’ exists. His simple use of the assertion that ‘there is in Jerusalem … a pool with five porticoes’ is best explained by the inference that he is writing before 66–70 ce, when the presence of the Herodian pool is a given and need not be explained; a post-70 ce readership would have assumed, given the destruction of Jerusalem, that the unexpected survival of a particular building complex required at least some qualification, however light. If the Gospel of John refers to the pool with five porticoes as simply existing, as standing, then the Gospel was most likely written prior to 66–70 ce.

4 The Consensus Concerning the Total Destruction of the City of Jerusalem in Flavian and Post-Flavian Literature, and the Absence of Post-70 ce Anachronisms in the Gospel

I have argued above that a) a grammatical analysis shows that the present tense ἔστιν in John 5.2 is a real, existential present, meaning that the colonnaded Pool of Bethzatha exists at the time the author is writing his Gospel, and b) a historical and archaeological analysis shows it is likely that the pool’s porticoes and the nearby Sheep Gate in the third wall were destroyed and/or dismantled during or at the end of the First Jewish Revolt, thus concluding that c) the Gospel was written before 66–70 ce.

Three alternatives remain, however, which all revolve around the assumption that the Gospel’s author might have been mistaken in stating that there ‘is’ in Jerusalem a colonnaded pool at the time of writing. The first alternative is that the author may have been using a pre-70 ce written source and copied it verbatim, without changing the present tense after 70 ce. The second alternative is that a post-70 ce author writing outside Judaea may not have known about the destruction of Jerusalem in general, or of the Bezetha area and its gates in particular, and may therefore have written about the colonnaded pool in the present tense. These two alternatives both concern an unintentional error. We will return to the third alternative below.

Certain authors have mentioned the first alternative of a post-70 ce author who forgot to change his source. In his Redating the New Testament (1976), Robinson drew attention to the fact that ‘John says not “was” but “is”’, but notes that ‘it is the only present tense in the context, and elsewhere (4.6; 11.18; 18.1; 19.41) he assimilates his topographical descriptions to the tense of the narrative’; Robinson concludes, ‘The natural inference […] is that he is writing when the building he describes is still standing.’ At the same time, Robinson was aware that ‘it is always open to the critic to attribute it to a source, which the evangelist has not bothered to correct’.Footnote ⁹⁶ This view that the present tense was used in the author’s pre-70 ce source had indeed been proposed by Antoine Duprez in 1970.Footnote ⁹⁷ In his 2007 commentary on John, Benedikt Schwank developed a similar view, although not with regard to a source, but with respect to the author’s own pre-70 ce draft of the Gospel. According to Schwank, the present tense in 5.2 indeed gives the impression that this sentence was written before the destruction of Jerusalem in 70 ce, and Schwank therefore takes the view that at least some parts of the Gospel reach back into the time before 70 ce. Yet without giving a reason, Schwank nevertheless finds that a completion of the entire Gospel before 70 ce is less convincing.Footnote ⁹⁸

I have not found the second alternative concerning the ignorance of a post-70 ce author outside Judaea, which prevented him from using the past tense as he should have done in scholarly literature, but it has been voiced in discussions. Both the first and the second alternative assume that the Gospel’s author made an unintentional error, either through a failure to update his pre-70 ce source or through ignorance of the situation in Jerusalem. According to both alternatives, he used the present tense erroneously but unintentionally.

The third alternative is that the Gospel’s author made an intentional error and wanted to give his readership the impression that he was writing before 70 ce, whereas in fact he was writing after the First Jewish Revolt. In this case, his use of the present tense is not merely erroneous, but deceptive. I have not found this alternative in the literature either, but it also came up in discussions.

The common feature of all three alternatives is that the Gospel’s author wrongly used the real, existential present to refer to the colonnaded pool – whether intentionally or unintentionally. In my view, the assumption that the author was mistaken in his use of the present tense and should have used the past tense is questionable, because our interpretation of the historical and archaeological evidence for the likely disappearance of the pool’s porticoes in 66–70 ce and the demolition of the Sheep Gate in the third wall around the Bezetha area in 70 ce indicates that the description of the colonnaded pool near the Sheep Gate is most naturally dated prior to the First Jewish Revolt. In my response to the first two alternatives, the claim that the author made an unintentional mistake either due to an oversight or due to his ignorance, I will argue that these options are unlikely because the destruction of Jerusalem was common and widespread knowledge after 70 ce, as demonstrated in Greek and Roman reports on post-70 ce Jerusalem. In my response to the third alternative, which asserts that the author made an intentional mistake because he wanted to mislead his readership, I will show that this is difficult to argue, as the Gospel contains no anachronisms to show that the pre-70 ce dating is contradicted by post-70 ce realities.

The enormous scale of Jerusalem’s destruction was indeed common knowledge after the First Jewish Revolt. Jews were very much aware of the destruction of Jerusalem in 70 ce, and Josephus himself says that after the destruction of the city, Vespasian farmed out Judaea, left Jerusalem destroyed, and did not found a new city in its place:

Furthermore, in the aftermath of the First Jewish Revolt, pagan Greco-Roman authors also had a complete awareness of the fact that Jerusalem has been utterly destroyed. In the first twenty-five years after its disastrous end, Roman authors of the Flavian era – historians and poets alike – clearly state that Jerusalem had been obliterated. Firstly, in his description of Judaea, Pliny the Elder states that Jerusalem is no longer one of Judaea’s toparchies (local government areas; Natural History 5.70), as Josephus still has it in his description of pre-70 ce Judaea (BJ 3.51–6).Footnote ⁹⁹ According to Pliny, Jerusalem is a ‘heap of ashes’ (5.73). Secondly, the same awareness that Jerusalem has been entirely destroyed is widespread among the Flavian poets, who often call Jerusalem ‘Solyma’.Footnote ¹⁰⁰ They refer to Jerusalem as ‘burned-out’ (Martial, Epigrams 7.55.7), characterise Titus as ‘blackened with the dust of Solyma’ (Valerius Flaccus, Argonautica 1.12–13) and envision visitors to Jerusalem as ‘tread[ing] Solyma’s ashes’ (Statius, Silvae 5.2.138–9). Subsequently, after the Flavian period, Greco-Roman authors continue to relate the destruction of Jerusalem in 70 ce, stating that ‘Vespasian destroyed it’ (Appian, Roman History 11, The Syrian Book (Syriaca) 252), that ‘the Roman Emperor’ razed the city of Jerusalem ‘to the ground’ (Pausanias, Description of Greece 8.16.5), that Jerusalem ‘was utterly destroyed’ (Cassius Dio 66.7.2 (=LCL 65.7.2)) and that Hadrian replaced it with a new city in 130 ce, ‘in place of the one which had been razed to the ground’ (69.12.1).

This all shows that a pervasive, empire-wide awareness that Jerusalem had been totally destroyed before Hadrian rebuilt it did exist after 70 ce. Therefore, it is very unlikely that the author of the Gospel of John would refer to the colonnaded Pool of Bethzatha in the present tense, as an existing structure in Jerusalem (5.2), if he were writing after 70 ce. No Jewish, Greek or Roman author would have referred to architectural features in Jerusalem in the present tense between 70 and 135 ce. A reference to a state of affairs in the present tense, if that state of affairs has changed, implies that this reference was made prior to that change. For this reason, the statement that there ‘is’ a monumental, colonnaded pool in Jerusalem was necessarily made before its likely destruction during the First Jewish Revolt.

The reason the Gospel’s author refers to the colonnaded pool in the present tense, rather than subsuming it in the past tenses of his narrative, is probably the landmark nature of this complex. In fact, in his entire Gospel, it is only here that he uses a present tense when he provides geographic and topographic descriptions, which he usually gives in the past tense. According to him, John the Baptist ‘was baptising at Aenon near Salim because water was (ἦν) abundant there’ (3.23); ‘Jacob’s well was (ἦν) there’, near the Samaritan city of Sychar (4.5–6); ‘there was (ἦν) a great deal of grass in the place’ at the Sea of Tiberias, where Jesus fed the five thousand (6.10); ‘Bethany was (ἦν) near Jerusalem’ (11.18); ‘there was (ἦν) a garden’ across the Kidron valley, where Jesus and his disciples went after their final supper (18.1); and ‘there was (ἦν) a garden in the place where he was crucified’ and buried (19.41–2). Yet only in John 5.2 does the author make an assertion about the current existence of a building in Jerusalem. We find parallels to this in two ethnographic assertions the author makes in the present tense concerning the relation between the Jews and the Samaritans (4.9) and Jewish burial customs (19.39–40). The author’s reference to the Pool of Bethzatha in the present tense, as a colonnaded complex that ‘is’ in existence, could easily have been evoked by the monumentality and recognisability of this complex in Jerusalem. After all, it was one of only two open-air pools that facilitated the pilgrims’ ritual cleansing on their journey to the Jerusalem temple, and the fact that it constituted a landmark explains why the author refers to it in the present tense.

This seems comparable to what we find in the Acts of the Apostles. When the author describes Paul’s transition from Asia Minor (Troas) to Macedonia (Philippi), he gives the narrative in the aorist and imperfect tenses, but when he describes Philippi, the setting for the narrative, he says that it ‘is (ἐστὶν) a leading city of the district of Macedonia and a Roman colony’ (Acts 16.11–12). Moreover, when the author relates Paul’s further journey from Philippi to Thessalonica, among the aorist tenses of his narrative, concerning Thessalonica he says that ‘there was (ἦν) a synagogue of the Jews’ (17.1), using the imperfect tense. Whereas the city of Philippi is a landmark in Macedonia, an unexceptional synagogue building in Thessalonica is not. The use of the present tense to describe the city of Philippi also seems to reflect the author’s assumption that, because of its landmark status, his readership might well be aware of Philippi’s existence, whereas he would assume that they are unaware of the existence of a Thessalonian synagogue, so that its description is easily absorbed into the past tenses of this narrative about the past. While a description of this synagogue in the past tense does not necessarily imply that it had ceased to exist at the time of writing, the present-tense description of the city of Philippi does refer to its existence at the time of writing. Thus, there is an asymmetry here: if a phenomenon described in the past tense in a narrative may either still have existed or ceased to exist at the time of writing, this does not mean that a phenomenon described in the present tense is equally ambiguous.Footnote ¹⁰¹ Rudolf Bultmann applied such a symmetry in a comparison between the present tense used in John 5.2 and the past tense used in John 11.18, which says that ‘Bethany was near Jerusalem, some two miles away’ (ἦν δὲ ἡ Βηθανία ἐγγὺς τῶν Ἱεροσολύμων ὡς ἀπὸ σταδίων δεκαπέντε). According to Bultmann, in John 5.2 ‘Ἔστιν […] hardly proves that at the time of the narrator the temple had not yet been destroyed; nor does ἦν 11.18 prove the contrary.’Footnote ¹⁰² This argument, which was still used by Köstenberger in his 2004 commentary,Footnote ¹⁰³ had already been proposed by Walter Bauer,Footnote ¹⁰⁴ to whom Udo Schnelle still refers in his 2016 commentary to support his judgement that ‘der präsentische Versanfang nicht besagt, dass Jerusalem und der Ort der folgenden Handlung z.Z. der joh. Vorlage noch bestanden’.Footnote ¹⁰⁵

Yet there is no symmetry here. Although one can describe something in the past tense without claiming that it has ceased to exist, the use of the present tense ‘is’ to describe something at the time of writing means that the author is making the assertion that the thing described indeed exists. Hence, in his description of one and the same journey taken in the past, the author of the Acts of the Apostles can say that ‘there was (ἦν) a synagogue of the Jews’ in Thessalonica (17.1), but that Philippi ‘is (ἐστὶν) a leading city of the district of Macedonia and a Roman colony’ (16.12). Similarly, among all the other places and locations in the Gospel of John, which the author describes in the imperfect tense in his narrative about the past, the author depicts only the colonnaded Pool of Bethzatha in the present tense, probably because it stands out as a monumental landmark in Jerusalem. If the author was writing after 70 ce, he would have known of Jerusalem’s utter destruction and, given the monumentality of the Pool of Bethzatha with its distinctive five porticoes, he would not have forgotten to change the present tense of his pre-70 ce written sources or his own pre-70 ce draft into the past tense. After the destruction of a similar urban landmark in September 2001, no American author would simply maintain pre-2001 sources or drafts saying, ‘There is in New York by the Hudson River the World Trade Center, which has Twin Towers’, and fail to change the present tense into the past tense.Footnote ¹⁰⁶

This precludes the two first alternatives, which are both based on the assumption of the author’s ignorance concerning Jerusalem’s fate post-70 ce. Given the scale of the destruction, no author writing outside Judaea could simply assume that such a large and monumental building as the colonnaded Pool of Bethzatha and such a significant landmark as the Sheep Gate were an exception to this state of affairs and were still standing after 70 ce. It would have been more natural for him to include the status of the colonnaded pool in the fate of Jerusalem as a whole and to describe it in the past tense, which would have blended with the other past tenses used in his narrative. This is what Nonnus of Panopolis did when he composed his Paraphrase of the Gospel of John in the mid-fifth century ce and rendered John 5.2 as follows: ‘There was (ἦν, imperf ind act 3rd sg) an artfully constructed bath within the well-watered sheep-gate, | under the five, long-sided porticoes of the edifice, | girded with the high-arched headband of well-wrought stones, | a wide bath (ἦν δέ τις εὐποίητος … εὐρυτενὴς ἀσάμινθος)’ (Book 5, lines 3–6).Footnote ¹⁰⁷ Yet instead of this more natural choice of the past tense, the Gospel’s author uses the present tense, which makes it more likely that he found reason to do so in the landmark character of this existing building and was thus writing prior to the events of 66–70 ce.

For the same reason, this author would not have left a present-tense verb in his written sources or his own pre-70 ce draft unchanged after 70 ce. It is important to note that John 5.2 is in fact a ‘narrative aside’, one of the many in the Gospel of John, which are inserted into the narration of the past, but constitute conscious contemporary communications between the narrator and the readers in a shared historical context, so that the present tense in John 5.2 is a deliberate choice, rather than an unrevised remnant of a source or previous draft.Footnote ¹⁰⁸ Even if John 21, the last chapter of the Gospel, was added by an editor who supplemented either the whole chapter (as a secondary ending to 20.30–1) or only the colophon at the end (21.24–5), this does not mean that the chapter itself must be dated after 70 ce, as the internal evidence of chapter 21 merely points to a date after Peter’s crucifixion in Rome in 64 ce, which is alluded to in 21.18–19 (cf. 13.36).Footnote ¹⁰⁹ Thus, the internal evidence of the entire Gospel points to a date of completion between 64 ce and the First Jewish Revolt (66–70 ce) – that is, to 65 ce.

The final, third alternative – that the author’s assertion that the colonnaded pool still exists was an intentional error, as he wanted to give the misleading impression that he was writing before 70 ce – cannot be proven on the basis of the Gospel’s internal evidence. The Gospel contains no anachronisms which necessitate that the author must have been writing after 70 ce. None of the anachronisms that have often been raised in support of this argument are cogent:

a) The expulsions from synagogues to which the Gospel refers (9.22; 12.42; 16.2), and which are often held to have only been possible on the basis of the rabbinic anti-heretic Birkat ha-Minim, were a practice already known to Paul (1 Thess 2.14–16; cf. 1 Cor 15.9; Gal 1.13; Phil 3.6). The dedicatory inscription in Greek from the synagogue of Theodotus (Θεόδοτος) in Jerusalem (1st cent. bce/ce, before 70 ce; Israel Museum, Jerusalem # IAA S 842; CIIP I no. 9), discovered near the Temple Mount,Footnote ¹¹⁰ and the tensions between the Greek inhabitants of Caesarea Maritima and the local synagogue in 66 ce, which Josephus records (BJ 2.285–92), indicate that the infrastructure and conditions for such frictions were already present prior to the First Jewish Revolt.

b) The absence of the Sadducees as a Jewish movement in the Gospel of John (and the presence of the Pharisees alone) is hardly indicative of a post-70 ce situation: in the Gospel of Mark – the earliest of the Synoptic Gospels, written around 70 ce – they only occur once (12.18), in marked contrast with the Pharisees. Martin Goodman has also warned that the assumption that Sadducees and other Jewish movements disappeared after 70 ce is an illusion.Footnote ¹¹¹

c) The assumption that addressing Jesus as ‘My Lord and my God’ (Ὁ κύριός μου καὶ ὁ θεός μου, 20.28) is anti-Domitianic, since it was Domitian who asked to be addressed as ‘our Lord (dominus)/Master (δεσπότης) and God (deus/θεός)’ (Martial, Epigrams 5.8.1, 7.34.8, 8.2.6, 9.66.3; Dio Chrysostom, Discourses 45.1; Suetonius, Lives of the Caesars 8.3, Domitian 13.2; Cassius Dio, Roman History 67.4.7, cf. 67.13.4), ignores the facts that the Septuagint is full of direct references to God as ‘Lord and God’ (see, e.g., Psalm 34 [35]:23 LXX: ὁ θεός μου καὶ ὁ κύριός μου) and that Philo of Alexandria also frequently uses the phrase ‘Lord and God’,Footnote ¹¹² let alone the frequent pre-Domitianic pagan references to God as ‘god and lord’, including Diodorus of Sicily’s reference to the Jewish god as the one who ‘is alone a god, and a lord of the universe’ (The Library of History 40.3.4). This way of referring to (the) god(s) as lord(s) continues in the Domitianic and post-Domitianic eras, but it did not arise then.

d) Finally, the Gospel contains no references to the destruction of Jerusalem in 70 ce. When the high-priestly factions in the Sanhedrin are concerned that ‘the Romans will come and take away our holy place and our nation’ (καὶ ἐλεύσονται οἱ Ῥωμαῖοι καὶ ἀροῦσιν ἡμῶν καὶ τὸν τόπον καὶ τὸ ἔθνος; 11.48), the emphatically placed ‘our’ suggests that they fear being deposed by the Roman procurators of Judaea (or in their absence, by the Roman governors of Syria), who, in the first period of direct Roman rule over Judaea (6–41 ce), were indeed appointing and deposing the Jewish high priests.Footnote ¹¹³ Therefore, there is no need to and no justification for translating this phrase as the NRSV does, saying that ‘the Romans will come and destroy both our holy place and our nation’. Never in the Gospel of John is there a clear post-70 ce reference to the destruction of the temple, and even in his narration of Jesus’ cleansing of the temple (2.14–22), the author refrains from suggesting that the destruction of the temple in 70 ce was a fulfilment of Jesus’ words concerning the destruction of the temple of his body (2.19, 21).

In the absence of such anachronisms, we must give the present tense in the formula employed in John 5.2 its full ‘existential’ force. Thus we reach the conclusion that, analogous to the standard Greek use of this formula, the Gospel’s internal evidence, combined with the likely destruction of the five porticoes of the Pool of Bethzatha in 66–70 ce and the demolition of the Sheep Gate in 70 ce, indicate that it was written when this colonnaded pool was still standing, before the First Jewish Revolt.Footnote ¹¹⁴