Luke’s Acts as a Historical Source for Paul

The Acts of the Apostles is critical to historians for establishing (a) the connection between Jesus and earliest Christianity, and (b) a chronology of the life of Paul and its relationship with his letters.  In this brief paper we will direct our attention to (b).

During the twentieth century, however, four criticisms have been directed against the usefulness to historians of the book of Acts for providing a historical and chronological basis for the life and ministry of Paul (see R. Riesner, Paul’s Early Period Grand Rapids:  Eerdmans, 1998, 3-28).

Four Criticisms of the Acts of the Apostles

(a)            As compared to Paul’s own references the Acts is not to be regarded as a ‘primary reference’ but as a ‘secondary reference’. This view is especially connected with J. Knox but has become critical orthodoxy for many. For some authorities Acts as a ‘secondary reference’ means that it is of little or no use to the historian, whereas for others it means that it is of use where it can be shown to agree with Paul.

(b)            Closely connected is the viewpoint that discovers historical divergences in Acts as an unreliable secondary source from Paul as the reliable primary source.  These include the omission by Acts of Paul’s sojourn in Arabia (Gal. 1:17) and its conflicting accounts of Paul’s first and second return visits to Jerusalem (Gal. 1:18-21/Acts 9:26-27; Gal. 2:1-10/Acts 11:27-30).

(c)            Passages in the Acts of the Apostles are historically inaccurate and significantly diminish the value of its text.  A prime example is Gamaliel’s speech to the Sanhedrin where he locates the revolutionary prophet Theudas before the insurrectionist Judas, contrary to the witness of Josephus (Acts 5:33-39). Josephus states that Theudas led his insurrection in the mid-40s, thirty years after the uprising of Judas the Galilean (Antiquities of the Jews, xviii.3; xx.97-99).  It is no less serious that the Theudas incident occurred between AD 44-46 whereas Luke quotes Gamaliel speaking to the Sanhedrin about Theudas in ca. 34, about twelve years earlier.

(d)            There is such theological disparity between the Acts of the Apostles and Paul’s letters that the two authors must have been unknown to each other.  It is claimed, for example, that the two authors do not share the same attitude to the law, and therefore to the centrality of the cross of Christ and the role of faith for divine justification.

Responses to these Criticisms

It is possible to make reasonable responses to these criticisms.

(a)            Two responses may be offered to the distinction between Paul as the ‘primary source’ and Luke’s Acts as the ‘secondary source’.  By ‘secondary source’ critics of Acts do not mean that Acts is directly derived from or dependent on the Pauline literary corpus (as Luke’s Gospel was directly derived from Mark’s Gospel).  In their view, to the contrary, Luke’s Acts depends on extraneous, late and unknown sources.

This brings me response (i) to this criticism.  It is that the ‘we’ and ‘us’ passages (Acts 16:10-16; 20:5-21:18; 27:1-28:16) are most sensibly understood as indicating the author’s presence alongside or near Paul during the five years those passages narrate. Significantly, these chapters are intensely more detailed than the preceding chapters of Luke-Acts prompting J. Fitzmyer to refer to a ‘a diary-like record’:

…they [details in Acts] are drawn from a diary-like record that the author of  Acts once kept and give evidence that he was for a time a companion of Paul (J. Fitzmyer, Luke the Theologian: Aspects of his Teaching  (London:Geoffrey Chapman, 1989, 22).

Understood in this way means that the ‘we’ and ‘us’ chapters should be read alongside Paul’s letters (insofar as the narratives overlap) as an equal primary source.  Paul’s letter to the Romans spells out his intention to come to Rome and his letter to the Philippians (most probably) written from Rome indicates that did in fact reach Rome.  Acts 27-28 authentically narrates why and how Paul travelled from Corinth via Judea to Rome.

Luke’s companionship with Paul AD 57-62 would have provided opportunity for the author of Acts to know about Paul’s life beforehand.  Through conversation and perhaps written memoirs Luke would come to know of Paul’s birth in Tarsus, his resettlement in Jerusalem, his conversion, his ‘unknown years’ between Damascus and Antioch, and his subsequent westward missionary journeys prior to their years of companionship.

Response (ii) is to point out that ‘primary source’ material isn’t necessarily free from bias and that ‘secondary source’ material isn’t necessarily inferior or inaccurate.  Who is to say, for example, that Paul did not underplay certain details in his memoir to the Galatians in the first two chapters of that letter?  This is not to say that he did, only that the possibility is there.  On the other hand, based on the hypothesis that Paul told Luke about his earlier life, is there any good reason to argue that he falsified the details at hand?  It is not doubted that he shaped his raw material, but that is not the same as arguing against his integrity overall or in matters of detail.

(b)            That Luke’s details vary from Paul’s at some points does not necessarily indicate that the author of Acts was ignorant of Paul’s missionary movements.  Such a hypothesis would suggest that Luke’s source for chapters 13-20 (as well as details of Paul’s life to that point) was remote from Paul, not dependent on him.

In fact, both the Acts (explicitly) and the Pauline corpus (implicitly) refer to the same theological-geographical ‘narrative’ for Paul.  Both writers interpret the promises of the Old Testament as confirmed in Christ and envisage the gospel message proceeding from Jerusalem to the gentile world.  More specifically, both Paul and Luke trace the spread of the gospel from Jerusalem in a westerly, Rome-ward direction.  That sense of direction emerges clearly from Paul’s Romans (chapters 15-16) and from the entire narrative of the book of Acts.

Luke was constrained by the capacity of his scroll and was forced to abbreviate and omit detail to fulfill his Jerusalem to Rome narrative.  This might explain why, for example, he passes over Paul’s story from Damascus to Antioch, a period of about fifteen years, in a few sentences and omits Paul’s journey to Arabia altogether.

The major problem identified by scholars is the disparity between Paul’s second return visit to Jerusalem narrated on the one hand by Paul (Gal. 2:1-10), and on the other by Luke (Acts 11:27-30).  According to Paul the purpose of the visit was to secure the pillar-apostles’ recognition of Paul’s proposed circumcision-free mission to the Gentiles, whereas Luke states that it was to deliver famine relief from Antioch.

It is right to ask, however, why should Luke’s account be treated as incorrect?  It is quite possible that Paul focused on the divisive issue of circumcision while passing over the delivery of famine relief (an in Luke’s narrative).  It is well known that Luke generally tends to play down divisions within the apostolic community whereas Paul was prepared to highlight them, which he does implicitly in Jerusalem and explicitly in the ‘Incident in Antioch’ (Gal. 2:1-10; 11-14) – especially when defending his doctrines to the Galatians, as he does throughout this letter.

Furthermore, Luke’s account of the beginnings of the westward missions from Antioch occurred immediately after Paul’s return from Jerusalem (Acts 13:1-3).  This is entirely consistent with Paul’s note that the Jerusalem ‘pillars’ agreed that Paul and Barnabas should ‘go’ to the Gentiles (Gal 2:9).  Their condition was that Barnabas and Paul were to ‘remember the poor’, which, Paul adds, was the ‘very thing I have taken pains also to do’ (Gal. 2:10 – As translated by E. de Witt Burton, Galatians ICC (Edinburgh: T & T Clark, 1980, 99.).  Paul’s retrospective defensive remark confirming Luke’s account of the journey from Antioch to Jerusalem to bring famine relief (Acts 11:27-30).

Many scholars, however, seek to eliminate Acts 11:27-30 as a genuine visit to Jerusalem and prefer to equate Gal. 2:1-10 with Acts 15:4-29 (otherwise known as the Jerusalem Council) as Visit 2.

There are substantial problems with this reconstruction.  One is that Visit 2 according to Galatians was specifically held ‘privately’ between Barnabas and Paul and James, Cephas and John (Gal. 2:2) whereas the Acts 15 meeting involved ‘the apostles and elders with the whole church’ (Acts 15:6, 22, 23).  Furthermore, the private meeting in Jerusalem (Visit 2) preceded the missions to the Gentiles (Gal. 2:9) and the plenary meeting in Jerusalem (Visit 3) succeeded the mission to the Gentiles, and was held to address the issues raised by the missions of Barnabas and Paul ‘among the Gentiles’ (Acts 15:12).

Galatians does not mention a Visit 3 to Jerusalem for the simple reason that it had not yet happened when Paul wrote the letter.  Paul wrote to the Galatians following the Incident at Antioch (Gal 2:11-14), which occurred after his return to the Syrian capital after his missions in Galatia.  It was only then that Barnabas and Paul travelled to Jerusalem for Visit 3, the Jerusalem Council.

(c)            What, then, can be said regarding passages in Acts that are regarded as historically inaccurate, in particular Gamaliel’s speech to the Sanhedrin where he locates the revolutionary prophet Theudas before the insurrectionist Judas?

It is possible that Gamaliel is referring to an otherwise unknown Theudas who preceded Judas.  Theudas is an abbreviation of Theodotus (‘gift of God’) that in turn is the Greek version of the Hebrew name, ‘Jonathon’.  Was Gamaliel referring to an insurrectionist named Theudas who arose during the time of Archelaus (3 BC-AD 6) or, before him, in the time of Herod (40-4 BC)?  Whilst this is theoretically possible it is unlikely because Gamaliel’s quoted words about Theudas and Judas closely match Josephus’ references to men of that name.

It appears, then, that Luke has reversed the true sequence of Judas and Theudas and placed words anachronistically in the mouth of Gamaliel.

In defence of Luke it is possible that the fault lay with the source or sources that Luke used.  My thesis is that Paul was a good source for Luke, based on their extensive companionship.  But for other events like the Gamaliel incident Luke depended on hearsay or written fragmentary chronicles.  It is not reasonable to fault Luke for matters about which he would have been dependent on hearsay or upon an earlier written account of that incident.

In any case, the reference to Gamaliel is but one problematic reference amongst many references to people cross-referenced in world history that are regarded as historically reliable.   These include the named members of the Annas dynasty (4:6), the famine that occurred in the days of Claudius (11:28), Herod the king (12:1), Sergius Paulus, proconsul of Cyprus (13:7), the ‘Politarchs’ of Thessalonica (17:8), the exile of Jews from Italy (18:2), Gallio, proconsul of Achaia, and the ‘Asiarchs’ of Ephesus (19:31).

Thus whilst candour requires acknowledgment of problems in the Gamaliel incident, this needs to be recognized in the broader context of many other unproblematic references.

(d)            It not come as a surprise that there is theological disparity between the Acts of the Apostles and Paul’s letters.  Luke was probably a Gentile and a God-fearer whereas Paul was a Jew, in fact a strict and educated Pharisee.

The pre-Christian Paul may have outwardly seemed ‘under law blameless’ (Phil. 3:6), but within his conscience he was aware – however dimly – that he was ‘a captive under law, a prisoner’ (Gal. 3:23-25), a Jew like other Jews ‘under a curse’ as a law-breaker (Gal. 3:10), in desperate need of divine redemption (Gal. 3:13; 4:4).  Given this circumstance it is understandable that Paul should write so passionately about the cross of Christ as God’s instrument of freedom, and of the role of faith not ‘works of the law’ (Gal. 2:21; 5:11; 6:14-15).

Nonetheless, there are echoes of Paul’s ‘righteousness’ language in Luke-Acts (Luke 18:9, 14; Acts 13:38-39).

Luke was already a disciple (from Antioch?) by ca. 50 when he joined Paul in Troas and travelled to Philippi where he remained for the next seven years (Acts 16:10; 20:6).  By the time he re-joined Paul in AD 57 he had doubtless formed his own theological views so that there is no reason to expect these to have been identical with Paul’s very distinctive theology.

Thus it is quite unreasonable to demand similarity of viewpoints between Paul and Luke and to argue that Luke could not have known Paul because these were not identical.


Our argument has been that the case against the historical value of Acts based on historical and theological divergences from Paul are significant but not ultimately sustainable.  The ‘primary’ versus ‘secondary’ viewpoint fails because a ‘primary’ source may be tendentious or forgetful and a ‘secondary’ source may be based accurately on the witness of the ‘primary source’.

Moreover, it is fallacious to require a ‘secondary’ source slavishly to follow the narrative of the ‘primary source’.  Both Paul and Luke follow a Jerusalem-to-Rome missionary thrust, but for his part Luke omits and compresses his narrative according to his overall literary-theological design.  Whilst the Gamaliel incident raises significant questions for Luke’s accuracy this issue must be seen within the broader context where his historical competence is demonstrable, especially in the ‘we’ and ‘us’ passages.

Furthermore, the demand that Luke’s theology must cohere tightly with Paul’s is unreasonable.  Paul was an intensely religious Jew and Luke was apparently a Gentile so that to expect an identical theological framework is unfair to both men.

Unimaginable Details in Acts 13-20

The data about Paul in the book of Acts is extensive, especially for the span of years between his persecutions and his final journey to Jerusalem, where the principal ‘we’ and ‘us’ passage begins.  Within that quarter of a century Luke narrates Paul’s movements and mission in considerable detail, especially the westward mission decade AD 47-57.

That extensive detail includes the names of people and places and the passage of time and these are too numerous to repeat.  Did Luke invent these details, as some suggest, so that these narratives should be regarded as fictional?  One has only to compare the Acts accounts with various later apocryphal works to see how unlikely this suggestion is.

For the moment let us consider a few unimaginable details, but details that have been confirmed through modern study.  One such example is the travel information related to the journey of Paul and Barnabas through Pamphylia, Pisidia and Lycaonia, the so-called ‘first missionary journey’ (Acts 13-14).  Paul and Barnabas passed through Perga and travelled directly to Antioch in Pisidia and from there to Iconium, Lystra and Derbe, whereupon the missionaries retraced their steps to Perga but departed for Antioch from Attalia.

What is not clear from these references is the nature of the cities and network of roads between them.  From modern scholarship including archaeology we have information about these cities and roads that most likely would not have been available to a writer who was inventing this narrative.  How could a fictional writer located elsewhere know that the relatives of Sergius Paulus Proconsul of Cyprus were significant in Pisidia, as demonstrated by the discovery of the Paulli inscription in Antioch?  This would explain why Paul was so keen to travel directly from Cyprus to Antioch, without preaching in the major city of Perga.  Could someone who invented these narratives know that Antioch, Iconium and Lystra were Roman colonies and thus relatively safe for Paul the Roman citizen to visit, and explain by he bypassed other major cities in that region?  Would a novelistic writer know that Antioch, Iconium and Lystra were connected by a network of well-made Roman roads, including the Via Sebaste, providing further reason why the missionaries preached in those cities?  Could pure invention explain why they travelled on a non-Roman road to obscure Derbe, except to escape the immediate danger from Lystra and Iconium?

Similar questions could be posed about Paul’s numerous other travel details, which modern scholars understand through easy access to research information but which would not have been apparent to an anonymous chronicler in antiquity who would have lacked access to maps and encyclopaedias to inject verisimilitude into contrived narratives.

It is more realistic in every way to attribute the travel and other information in Acts 13-20 to Paul himself who, in turn passed it on to Luke, whether orally or by writing or both.

But this brings us again to the significance of the ‘we’ and ‘us’ passages in the Acts of the Apostles.

The Plausibility of the ‘we’ and ‘us’ Hypothesis

Our argument is that the unimaginable and otherwise inexplicable details in Acts 13-20 are best understood as originating directly from Paul to Luke, who then wove them into his global narrative in Luke-Acts.

The significance of the ‘we’ and ‘us’ narratives, especially in Acts 21-28, is that they directly connect Luke as Paul’s companion and for no less than five years.  The overwhelming probability is that Luke became acquainted with Paul’s earlier life, including his missionary travels, through those years of companionship.  Although many scholars dispute the ‘we’ and ‘us’ passages as pointing to this, Martin Hengel is clear on this point.

…the remarks in the first person plural refer to the author himself.  They do not go back to an earlier independent source, nor are they merely a literary convention, giving the impression that the author was an eyewitness… ‘We’ therefore appears in the travel narratives because Luke simply wanted to  indicate that he was there (Martin Hengel, Acts and the History of Earliest Christianity, London: SCM, 1979 ET, 66).

Such a conclusion is straightforward and sensible.  If indeed true it undergirds the historical integrity of the greater part of the book of Acts.  Without that integrity, as indicated earlier, it would not be possible to identify the connection between Jesus of Nazareth and earliest Christianity, or to provide any kind of framework for the missionary career of Paul and the dispatch of his letters to the churches of his mission.

Although written many years ago, the verdict of Alfred Plummer continues to be applicable.

It is perhaps no exaggeration to say that nothing in biblical criticism is more             important than this statement’ – ‘The Author of Acts was a companion of S. Paul’ (Alfred Plummer, St Luke ICC (Edinburgh: T&T Clark, 1901, xii.)


Although the various criticisms of Acts appear to damage the credibility of that text for commentary on Paul, those criticisms diminish when carefully evaluated.  The ’we’ and ‘us passages in Acts 27-28 are most cogently understood as the work of a companion of Paul throughout the five years, AD 57-62.  Such a companionship would equip one who was to write about Paul’s earlier years, especially the decade of westward mission, AD 47-57.  The alternative is that such narratives were essentially invented and therefore novelistic.  However, the numerous details of Paul’s journeys narrated in Acts, which are corroborated through modern research, would have been unimaginable to the writer of a contrived chronicle.