15 December 2023

The Westminster Divines and the Alexandrian Codex

By Stephen Steele

Stephen Steele is minister of Stranraer RP Church in Scotland. He has an MA from Queen’s University Belfast where his focus was on nineteenth-century Presbyterianism.


It has been assumed by those on both sides of the ‘Textus Receptus’ debate that the Westminster Divines did not have access to any of the Alexandrian manuscripts which later saw the dominance of the ‘Received Text’ overturned in critical editions of the Greek New Testament from the nineteenth century onwards. This article shows that, contrary to these assumptions, some Westminster Divines made use of a key Alexandrian manuscript which was gifted to England sixteen years after the publication of the KJV and originally intended for King James himself. Although Codex Alexandrinus was not published until 1786, various Westminster Divines had access to either the manuscript itself or collations of its readings. It is extensively cited in the ‘Westminster Annotations’ (a Bible commentary commissioned by the same Parliament that summoned the Westminster Assembly), while leading Westminster Divine Thomas Goodwin preferred its readings to the TR in a number of places in his published Works. The enthusiastic – and uncontroversial – use of this new manuscript by these Divines is one strand of evidence that, contrary to modern claims, Westminster Confession of Faith (and London Baptist Confession) 1:8 do not require the use of the Textus Receptus.


A key strand in the critique of modern Bible versions by some in the evangelical and Reformed world is the alleged superiority of the ‘Textus Receptus’ (TR) – the family of printed Greek editions which mostly lie behind the King James Version (and the New King James Version) – over against modern critical Greek texts.

This ‘TR-only’ position has been reinvigorated in some Presbyterian and Reformed Baptist circles in recent years by the emergence of ‘Confessional Bibliology’. This movement claims that subscription to the Westminster Confession of Faith or Second London Baptist Confession must include adherence to the ‘Textus Receptus’.[1]

After all, many of the Greek manuscripts used by modern Bible translations were not available to the TR editors in the sixteenth and seventeenth centuries. Since the Confessions teach that God’s word has been ‘kept pure in all ages’ (1:8), surely more recently discovered manuscripts must be rejected where they differ from the TR – even if they are much older than those available to the TR editors.

But does ‘Confessional Bibliology’ actually represent the position of those who framed the Westminster Confession? How would the Westminster Divines have reacted to new discoveries of Greek manuscripts which only became available to scholars after the King James Bible was published? And what if those manuscripts had originally come from Alexandria in Egypt of all places?

The Divines, we are told, would have resisted all efforts to correct the received text based on discoveries of manuscripts which were much older than those available to Erasmus and the other TR editors. They would have rejected them as suspect – if not Satanic.[2] We are told that B. B. Warfield’s idea that ‘the autographic text of Scripture must be sought in all manuscripts, including those recently discovered’ was ‘a foundational departure from the views of Reformed orthodoxy and the Westminster Assembly’.[3]

Those on both sides have tended to assume, however, that the question of what the Divines would have done with newly discovered manuscripts is purely hypothetical. After all, an English Bible translation based on more recently discovered manuscripts – the Revised Version – was not published until 1881. The typical narrative, as taken from a recent book arguing for the TR-only position, is that ‘with the discovery of older manuscripts in the nineteenth century, textual critics were motivated to produce new, reconstructed editions of the Greek New Testament’.[4]

We are told by a leading Confessional Bibliologist that Westminster Divines like Daniel Featley, ‘do not teach that some of the Word of God was missing, yet to be located in long lost manuscripts’.[5] Whether or not that is an accurate summary of the Critical Text position, the claim being made is that the Divines would have seen no need for new manuscripts – because they believed that God had preserved every jot and tittle of his word in the Textus Receptus.

Those on the other side of the debate say that it is unfair to hold the Divines up as TR-onlyists, when that was the only Greek text they had access to.

However the works of a leading Westminster Divine – as well as the Westminster Annotations (a Bible commentary that various Westminster Divines contributed to) – actually show us Westminster theologians using a newly discovered manuscript in preference to the TR in places.

Part 1: Thomas Goodwin

Thomas Goodwin (1600-1680) delivered approximately 400 speeches at the Westminster Assembly, was appointed to 35 committees, ‘respected by the Scottish commissioners’ and given oversight of printing of the assembly’s papers. While he is sometimes described as the first congregationalist, ‘Goodwin is more accurately remembered as one of the last of the puritans’.[6]

Goodwin was, additionally, one of those appointed by the Puritan-dominated Long Parliament to oversee the revision of the King James Bible in 1653 – something that had been called for by Westminster Divine John Lightfoot in a sermon to the House of Commons in 1645.[7] Others involved in the project included fellow Westminster Divines Joseph Caryl and William Greenhill, along with John Owen.[8] The attempted revision did not, however, survive the death of Oliver Cromwell and the restoration of Charles II.

In a posthumously published work, The Glory of the Gospel, Goodwin is enthusiastic both about the theoretical and actual discovery of ancient Greek manuscripts. Commenting on Colossians 1:26 – ‘Which hath been hidden from ages and from generations’ – Goodwin gives the following illustration:

To have an old copy of the New Testament, though it doth not differ three words throughout the whole from what we commonly have, yet if it be an old copy (as lately one of the Septuagint, written thirteen hundred years ago, was sent over), what a value is there set upon it![9]

He also uses the illustration of scholars finding manuscripts by a church father, the rediscovery of the Book of Enoch quoted by Jude – rumoured to have been found in Goodwin’s day, but not actually rediscovered until 1773 – and the discovery of Solomon’s writings on herbs and plants.

The discovery of an ancient manuscript of the Septuagint was no small thing, particularly with regard to discussions about the authority of the Hebrew Bible. Indeed, ‘Protestant scholars felt that Codex Alexandrinus could counter the Catholic possession of the other famous codex of the Septuagint, the Vaticanus’.[10]

However, as the editor of Goodwin’s Works notes, what the Puritan at that point in his life thought to be only an old copy of the Septuagint (the Greek translation of the Old Testament), actually turned out to contain the New Testament as well. Goodwin’s editor says that the author is ‘Doubtless’ referring to ‘the famous Alexandrian manuscript, which was sent from Constantinople, as a present to Charles I., in 1628’. Today we know it as Codex Alexandrinus – an almost complete copy of the New Testament from the fifth century. It was sent to England by the Calvinist patriarch of Constantinople sixteen years after the publication of the King James Bible, and originally intended for King James himself.[11]

Along with Codex Sinaiticus, Codex Vaticanus and Codex Ephraemi Rescriptus, it is one of only four manuscripts from the first millennium which were originally whole Bibles. In the gospels it is the oldest example of the ‘Byzantine’ text, but in the rest of the New Testament ‘it ranks along with Vaticanus and Sinaiticus as representative of the Alexandrian type of text’.[12] It has the distinction of being ‘the first manuscript of great importance and antiquity of which any extensive use was made by textual critics’.[13]

Of course, Goodwin referring to Codex Alexandrinus is one thing. But would he have used it to correct the received text? When the Westminster Divines said that the Greek New Testament had been ‘kept pure in all ages’, did they mean that the Textus Receptus (first published 1516, at which point it lacked the Comma Johanneum) could not be questioned?

After all, does Goodwin not use the example of a hypothetical New Testament manuscript that ‘doth not differ three words throughout the whole from what we commonly have’. But what if it did? Would he have rejected it as untrustworthy? And would he have considered the mention of textual variants from the pulpit as endangering his hearers’ faith?

In fact, as becomes clear in reference to a number of different Biblical texts, Goodwin had no qualms about suggesting there were places where the true text of the New Testament was preserved in an Alexandrian manuscript, rather than in the received text.

After all, as Goodwin explained in a sermon, ‘There are variæ lectiones of the New Testament, as well as of the Old; that is, various readings’.[14] In other words, all Greek manuscripts of any significant length vary from one another. Any printed Greek text must choose one reading over another (though will usually list alternatives in the ‘apparatus’). Since infallibility is not promised to the editor of any printed Greek edition, Goodwin had no qualms about opting for readings which are today found in the modern Critical Text, over against either the Textus Receptus or the Majority Text.

Goodwin explicitly appeals to Codex Alexandrinus against the TR on at least two occasions (comprising five variants). On another two occasions he opts for (or considers) readings found in Alexandrinus (and not the TR) without reference to it. Or perhaps without knowledge of it – Goodwin may only have had access to readings from Alexandrinus through the publication of Bishop Brian Walton’s London Polyglot in 1657.[15]

As we will see below, fellow Westminster Divine Daniel Featley had access to Alexandrinus in the 1640s. In 1650, Edward Leigh, a lay theologian who served as a teller at the Westminster Assembly, referred to Alexandrinus to argue that the postscripts to some of the epistles in the Textus Receptus were not original. He described the Codex as ‘the most ancient Parchment Manuscript Greeke Copy of the Bible, which Mr. Patrick Young hath to publish’.[16] A 1651 letter from Archbishop James Ussher (who turned down an invitation to participate in the Westminster Assembly) mentioned ‘the Alexandrian copy…which [Young] intendeth shortly to make publick, Mr. [John] Selden [a lay member of the Westminster Assembly] and myself every day pressing him to the work’.[17] Young, the Royal Librarian, failed to achieve his goal of publishing the entire manuscript, but collated Alexandrinus for Walton’s Polyglot.[18]

Whether Goodwin’s access to Alexandrinus was to the manuscript itself, or via other means, the fact that he opts for some of its readings over the TR is very significant. While a handful of appeals to a newly discovered manuscript may not seem overly significant, TR-onlyism, by definition, must reject any reading that didn’t make it into (certain) printed TR editions. For them, to admit that the TR is less than jot and tittle perfect, or to accept any non-TR reading, would be akin to giving up the epistemological foundation of the faith, and to deny that the Bible has been kept pure in every age.

This, we are told, was the position of the Westminster Divines. As will be demonstrate in this section, it was not the position of leading Westminster Divine Thomas Goodwin.

1 Peter 5:10

‘The God of all grace…make’ – KJV (TR and Majority Text)

‘The God of all grace…will’ – ESV (Modern Critical Text)

As Goodwin explains, the question here is ‘Whether these words be a prayer of the apostle’s unto God, or a direct promise from God?’. The difference, as Goodwin elaborates on in a footnote, is one Greek letter in each of four words affected, for example: καταρτίσαι v. καταρτίσει.

He finds support for reading this as a promise from a number of manuscripts, including ‘the ancient manuscript sent by Cyril into England’, i.e. Codex Alexandrinus.

Goodwin then spends 1600 words arguing against defenders of the TR reading on the basis of internal evidence. He cites a number of other Scriptural passages where the same words used in 1 Peter 5 take the form of promises, and concludes: ‘why therefore should not those copies that make it so here be esteemed genuine?’[19]

Why not indeed? Goodwin had no a priori commitment to the TR.

Acts 11:20

‘Hellenists’ – NKJV, ESV, NRSV (TR, Majority Text, Westcott & Hort, Modern Critical Text)[20]

‘Greeks’ – Revised Version, NIV, NASB, CSB (Griesbach, Wordsworth, Lachmann, Tischendorf, Tregelles)

Goodwin explains: ‘There hath been a question among some interpreters, whether these Grecians to whom these Jews preached were of Grecian birth and race, or Jews by race… called Ἑλληνισταὶ, or Grecising Jews’.

As Goodwin points out, the context strongly suggests Gentiles as opposed to Jews. Given that context, he opts for the reading of Alexandrinus over against the TR and the vast majority of Greek manuscripts: ‘The opposition clearly carries it; so accordingly in the manuscript copy sent by Cyril, that worthy patriarch of Constantinople, to king Charles I., they are expressly called (as it is here translated) Ἑλληνὲς, Grecians by birth and extraction’.[21]

Goodwin not only argues against the TR, he also goes against the Majority Text and the reading found in the Modern Critical Text (though critical texts in the past have agreed with Goodwin).

The point is not that Goodwin always agrees with the Modern Critical text, but that he is willing to accept the reading of a newly available Alexandrian manuscript in place of the TR.

H. A. Scrivener (the nineteenth century editor of the TR edition published by the Trinitarian Bible Society – but who did not hold to the TR-only position) notes his disagreement with the TR here and agrees with Goodwin.[22]

Jude 1:1

‘sanctified by God the Father’ – KJV (TR and Majority Text)

‘beloved in God the Father’ – ESV (Modern Critical Text)

Goodwin explains: ‘You have it indeed here read, and translated, “Sanctified by God the Father;” but if we consult both commentators and Greek original copies, as they are also cited by interpreters, we shall find that diverse, as authentic copies, as those that read it sanctified‚ &c., do write it beloved, in‚ or of, or by God the Father, ἠγαπημενοις, beloved, instead of ἠγίασμένοις.

(Although Goodwin doesn’t explicitly cite Alexandrinus, it reads ‘beloved’ – as do the two codexes particularly disparaged by TR-onlyists, Sinaiticus and Vaticanus).

Goodwin then explains what he considers the proper procedure for deciding between the two options (significantly, it doesn’t involve going with the TR no matter what). In cases ‘where there are found two such readings in so many copies ancient, and but a small difference in the Greek words themselves, which might easily occasion a mistake in the writers’, the question is to be decided by the context and comparing Scripture with Scripture (i.e. internal evidence).

Having weighed the evidence, Goodwin opts for ‘beloved’. Even before coming across this alternate reading, Goodwin says that something had always seemed out of place to him with the TR reading.[23] But now, having come across an alternative that fits better, he is convinced that it ‘hath far the advantage and appearance for it, to have originally fallen rather from our apostle’s pen’.[24]

Goodwin’s big concern is to discover what the Apostle originally wrote – and in pursuit of that desire he is willing to go outside the TR tradition.

Ephesians 1:18

‘the eyes of your understanding’ – KJV (and TR)

‘the eyes of your heart’ – ESV (Majority Text and Modern Critical Text)

This final example is significant – not because Goodwin comes down strongly on one side or the other – but because the TR and the Majority Text are often conflated. In fact, they differ in places – and where they do, TR-defenders are forced by their a priori assumptions to defend the TR reading over against the Byzantine one, even when there is little or no Greek manuscript evidence for it.

In this case, as the Lutheran Johann Albrecht Bengel noted in the first half of the 1700s: ‘Rec. Text, without any of the oldest authorities, reads διανοίας, of the understanding’. However, as he goes on to note, Alexandrinus, Vaticanus and many others support ‘heart’.[25]

As Princeton theologian Charles Hodge put it: ‘Instead of διανοίας understanding, the great majority of ancient manuscripts and versions read καρδίας heart which is no doubt the true reading’.[26]

Goodwin cites ‘the king of Spain’s Bible’ [ie the Antwerp Polyglot] in favour of ‘heart’ but then goes on to argue that there is not much difference between the two readings, concluding: ‘I speak this to reconcile those diverse readings which the copies have’.[27]

A TR-onlyist, however, would not have the option of reconciliation – non-TR readings must be rejected. Goodwin was not a TR-onlyist.

Goodwin Summary

In conclusion then, Goodwin explicitly cites Alexandrinus twice, both times agreeing with its readings over against the TR. On another occasion, he goes with the reading of Alexandrinus against the TR, without explicitly citing it. On a final occasion, he attempts to reconcile an Alexandrian (and Majority) reading and a TR reading. He has no a priori commitment to the TR. On 3 occasions, Goodwin opts for (or considers) a reading which we now know is also in Sinaiticus and Vaticanus.

A leading Confessional Bibliologist claims Goodwin for the TR-only position by citing his remark that when we read any epistle, ‘the whole weight of their apostolical spirit and authority in them is to fall upon all our consciences and spirits…to assure our hearts of the unerring truth of every tittle of them’.[28] Goodwin however says this while commenting on the very same chapter of Ephesians in which he would later cite a non-TR reading as potentially original. Given this is the same sermon series where he notes the reality that there are ‘various readings’ of the New Testament, Goodwin obviously believed that the first task of the expositor was to try and ascertain which reading is original. The only tittles which have ‘apostolic spirit and authority’ are those actually written by the Apostles – not simply any which happened to make it into the TR.

Significantly, Goodwin cannot even be claimed as a Majority Text advocate. Where the Modern Critical Text and the Majority Text disagree (two of the four examples), Goodwin opts for a reading now found in the Critical Text. Nowhere does he argue for a reading based merely on counting manuscripts. After Alexandrinus became available, it seems that the oldest and best manuscript he had access to took priority. As the KJV translator William Eyre wrote to Archbishop James Ussher in 1607: ‘The reading of a passage of the New Testament which can be proved by the authority and reliability of older and more correct copies, though they may perhaps be fewer in number, is (other things being equal) to be preferred’.[29]

Part 2: The Westminster Annotations

The Geneva Bible, beloved of the Puritans, was famously disliked by King James I because of its marginal notes. James declared ‘I profess I could never yet see a Bible well translated in English; but I think that, of all, that of Geneva is the worst’.[30] When laying down instructions for a new version to replace it, James specifically ordered that ‘no marginal notes at all [are] to be affixed’.

When the new version (the KJV) was completed in 1611, one of the reasons for its initial unpopularity was its lack of notes. ‘The people complained that they could not see into the sense of the Scripture, so well as formerly they did, by the Geneva Bibles, because their spectacles of Annotations were not fitted to the understanding of the new Text, nor any other supplyed in their stead’. As ‘the reading public had grown accustomed to having at their disposal scriptural aids for both private study and devotions’, they ‘therefore continued to use the Geneva Bible at home’, despite the printing of it being banned from 1616.

With the collapse of press censorship in 1640, a request was made to print the Geneva Bible notes (or a revision of them) in the margins of the King James Version. The theologians commissioned to the work by parliament soon found that the margins of the text were too confining, and it was agreed that the annotations should constitute a separate volume. The result was a work that became known as the ‘English’ (to distinguish it from Dutch and Italian equivalents) or ‘Westminster’ Annotations. They were ‘the product of some of the best Reformed minds in England’ and the preface to Matthew Poole’s Annotations, a quarter of a century later, testifies to their popularity. The first edition was published in 1645, the same year the Westminster Assembly began work on the Confession of Faith, with a much-expanded third edition coming in 1657. This final edition ‘became the closest thing to a comprehensive Bible commentary that English readers had yet seen’.

Although the Annotations ‘have been little studied, whether historically, exegetically or theologically’, their importance for understanding the theology of the Westminster Confession of Faith should not be underestimated. As Richard Muller has pointed out, there were ‘two sets of normative documents commissioned by the Long Parliament, namely, first, the exegetical and interpretive conclusions embodied, for public use, in the Annotations…and second, the doctrinal standards embodied as the Westminster Confession of Faith and its accompanying catechisms’. Even though there was no official relationship between the Annotations and the Westminster Standards, the former ‘provide a highly proximate index to the understanding of Scripture behind the doctrinal definitions and the biblical proofs found in the confession and catechisms’.[31] Half of the contributors (John Ley, William Gouge, Francis Taylor, Edward Reynolds, Thomas Gataker and Daniel Featley) were Westminster Divines.[32] Neither of the two studies on the Annotations (by Lampros in 1995 and Muller in 2007) make any reference to the fact that its pages repeatedly cite Codex Alexandrinus.

‘The most ancient Copie of Tecla’

The very title of the Westminster Annotations – a Bible commentary aimed at ordinary Christians – highlights the fact that the Biblical manuscripts which have come down to us all differ from one another. The same title that advertised ‘doubts resolved’ also advertised ‘various readings observed’ – apparently without any concern that these two goals might be in conflict. The Annotations reference various translations into other languages – such as Septuagint and the Syriac – but of particular significance for our purposes is its appeal to ‘the most ancient Copie of Tecla’.[33] Tecla (or Thecla) was ‘a fourth-century Egyptian martyr, often cited as the scribe for the Codex Alexandrinus’.[34]

The manuscript she was believed to have copied out was sent to England, as noted above, in 1628. It was regarded as ‘a jewell of anitiquity not fit to be kept among Infidels’.[35] Brian Walton, who used it for his Polyglot in 1657, described it as ‘one of the Noblest MSS. in the world…written in Capital Letters, without accents, or distinction either of words or sentences, which kind of writing hath been out of use for above a thousand years’.[36] It was kept in the Royal Library at St. James’s Palace.[37]

As noted earlier, the expectation by some of those present at the Westminster Assembly that this ancient manuscript would soon be published was disappointed. (This would be a recurring theme. After the Restoration, Charles II refused to grant permission for it to be engraved on copper plates, despite his librarian telling him that ‘it would appear glorious in history, after his Majesty’s death’. The King replied, ‘Pish, I care not what they say of me in history when I am dead’.[38] It would be more than 150 years after Alexandrinus’ arrival in England that its Greek text of the New Testament would finally be published, by Carl Gottfried Woide, in 1786.[39]

There was at least one Westminster Divine however who had access to Codex Alexandrinus way back in the 1640s. The Westminster Annotations on the Pauline Epistles, which explicitly reference ‘Tecla’ 118 times, were written by Daniel Featley. Featley was both a Westminster Divine and King James Bible translator, serving as part of the First Oxford Company.[40] He was ‘the only leading Episcopalian’ to accept a seat at the Westminster Assembly.[41] In his time there Featley ‘was one of the vociferous advocates for the Imputation of the Active Obedience of Christ’.[42] His time at Westminster came to an abrupt end when he was caught in a ‘sting’ operation, expelled from the Assembly and imprisoned by the House of Commons for allegedly being a royalist spy. However ‘even after imprisonment Featley continued to serve the Puritan cause’ and ‘at Parliament’s request he wrote a learned treatise against Roman Catholicism, for which he was allowed three books at a time from his library’.[43] From prison he also wrote his most famous work, The Dippers Dipt, against the Anabaptists. It was published in 1645 – the year the first edition of the Westminster Annotations were published – and the final year of Featley’s life. Edmund Calamy (son of the Westminster Divine of the same name) tells us that the annotations on St Paul’s Epistles were by Dr Featley, but ‘broken and imperfect on the Account of the Author’s dying before he had revis’d or finish’d them’.[44] Featley was esteemed by most to be ‘one of the most resolute and victorious champions of the reformed Protestant religion of any of his time; a most smart scourge of the church of Rome; a compendium of the learned tongues, and of all the liberal arts and sciences; and though of small stature, yet he had a great soul, and learning of all kinds compacted in him’.[45]

Lest any would downplay the Annotations’ use of an Alexandrian manuscript simply because Featley was an Episcopalian, it should be remembered that the annotators were deliberately left anonymous ‘despite the relative eminence of some of them as theologians and exegetes’. This was because the intention ‘had not been to produce an original work, but rather standard commentary that drew on the already sizeable and significant Reformed exegetical tradition’. As Muller has noted, ‘the impression that one receives from the prefaces to the Annotations, moreover, is of a corporate endeavour in which the names of the individual annotators were not to be highlighted in any way’.[46] Some of the contributors remain unknown.

The first reference to Alexandrinus in the Annotations (usually cited as ‘Gr. Tec.’) comes at Romans 1:8. Overall, Featley references Alexandrinus 44 times in his annotations on Romans. It is cited at least once in every Pauline epistle as follows: 1 Corinthians (27), 2 Corinthians (11), Galatians (7), Ephesians (11), Philippians (6), Colossians (4), 1 Timothy (3) and once each for 1 and 2 Thessalonians, 2 Timothy, Titus and Philemon. While Richard Muller leaves open the possibility that Featley also contributed the notes to ‘The Epistle of Paul to the Hebrews’, the absence of any reference to Tecla there, as well the fact that the notes on Hebrews were expanded in future editions (after Featley’s death), means we can confidently say that he did not.[47]

How did the Annotations use Alexandrinus?

The vast majority of times Alexandrinus is cited, it is done so without comment. In other words, Featley does not usually tell us whether he prefers the reading of the Textus Receptus, or that of Alexandrinus. The title of the Annotations had advertised ‘various readings observed’, and most often the reading of Alexandrinus is simply observed. Just because Featley quotes a reading from Alexandrinus does not necessarily mean he agrees with it. Presumably however the variants listed are ones he considered viable. Only twice does he explicitly accept or reject a reading from Alexandrinus. The reading he rejects is also rejected by both the Textus Receptus and the Critical Text. The reading he accepts is, in today’s language, an acceptance of the Critical Text over the Textus Receptus.

Is the Westminster Annotations’ use of Alexandrinus insignificant therefore, since on most occasions the reading is just cited without a judgement as to whether it is correct or not? Not at all. Leading TR-onlyists claim that the Textus Receptus provides us with the ‘perfect’, ‘preserved’, ‘certain’, ‘stable’, ‘settled’, ‘not changing’, ‘completed’, ‘agreed upon’, ‘fixed’ text of Scripture. If those involved in producing the Westminster Annotations believed those adjectives were true of the Textus Receptus, there would have been no point in listing alternative readings. Furthermore, TR-onlyists tell us that modern versions are questioning the Word of God simply by including footnotes. ‘The footnotes of many Bibles, including the New King James Version, actually undermine the unchanging nature of the text of Scripture to the common reader’.[48] By that standard, the Westminster Annotations – and the KJV itself which has a textual note saying that Luke 17:36 is missing from most manuscripts – are undermining the unchanging nature of the text of Scripture to the common reader. But perhaps they are just being honest.

The mere listing of readings from Alexandrinus is also significant because, to take Romans as a sample, out of 44 citations, around half make it into the Modern Critical Text. To extrapolate that out, of the 118 citations of Alexandrinus in the Westminster Annotations, it is likely that at least 50 make into Nestle-Aland’s 28th edition of the Greek New Testament (published in 2012). Almost all the Alexandrinus readings which make it into the Modern Critical Text also made it into Westcott and Hort’s New Testament in the Original Greek in 1881. So despite all the advances in textual criticism since the Annotations were published and despite all the new manuscripts that have been discovered, a straight line can be drawn from half of the ‘various readings observed’ in a single manuscript by a Westminster Divine in the 1640s right through to the latest scholarly edition of the Greek New Testament in use today.

How significant are the variants simply cited by the Annotations? Particularly those that make it into the Modern Critical Text? The answer is that they are like most viable textual variants – not very significant (despite leading ‘Confessional Bibliologists’ claims that the Critical Text is ‘a completely different underlying text’ or ‘a radically different text’).[49] A couple of the variants do not even necessarily affect translation – such as περὶ (‘about’, ‘concerning’, etc) instead of υπερ (‘for’, ‘on behalf of’, ‘because of’, etc) in Romans 1:8. Another variant highlighted in the Annotations is simply a difference in word order: ‘indignation and wrath’ v. ‘wrath and indignation’ (Romans 2:8). Four of the variants are the difference between γαρ (‘for’) and δε (‘but’) – followed by the Critical Text in Romans 4:15 and 15:8, but not in 8:18 or 8:22. Another is whether ‘infirmity’ in Romans 8:26 should be singular (as per the CT) or plural (as the TR). The question at Romans 9:32 is whether Paul writes ‘works of the law’ as he does many times elsewhere – or if he simply wrote ‘works’ and later scribes conformed it to the other passages. In Romans 10:15 the words translated ‘the gospel of peace’ are in the KJV but not modern translations; Alexandrinus does not contain them, and modern scholars believe that they were probably inserted to make the citation correspond more fully to the Septuagint.[50] Of course, whether something is an addition or omission depends on what you believe the standard is. If Alexandrinus (and the Modern Critical Text) represent the original reading of Romans 11:22, then the Textus Receptus removes the word ‘God’. Likewise, the TR omits the word ‘Jesus’ in Romans 8:34. On the other side of the coin, Featley is unafraid of listing variants that are potentially less ‘orthodox’ than the TR – such as in Romans 15:19 when Alexandrinus has ‘Holy Spirit’ rather than ‘Spirit of God’ (NA28 and THGNT both include ‘God’ – though NA28 does so in square brackets).[51] Incidentally, in all the examples above, Alexandrinus stands against not just the TR but also the Majority Text.

What of the variants which are discussed at more length? Romans 12:11 is significant because it is an occasion where different TR editions disagree with each other. The TR edited by Robert Estienne (Stephanus) in 1550 has ‘serving the time’, whereas that edited by Theodore Beza (1598) has ‘serving the Lord’. The KJV translators, as they did on 110 other occasions, went with Beza over against Stephanus (they did the opposite 59 times and differed from both 67 times).[52] (The ‘Textus Receptus’ that TR-onlyists use today is a reverse-engineered Greek Text produced in 1881, based on the text critical choices of an English translation, the KJV). Featley notes that ‘In some editions it is…serving the time: which reading if we should admit it, we must not understand the Apostle as if he commanded us to be temporizers’. ‘But’, he continues, ‘in the most ancient Copie of Tecla, and generally in the most correct Editions, the word is not καιρω, but κυρίῳ, not the time, but the Lord’.

Just as the various TR editions differ from one another at times, so do the various critical texts. A good example of this is at 1 Corinthians 2:1, where three Greek letters make the difference between ‘testimony’ of God or ‘mystery’ of God. Critical texts differ: Tregelles and THGNT (along with the TR and MT) opt for μαρτύριον, while WH and NA28 prefer μυστήριον. Featley gives both options: ‘That is, the Gospel which is the testimonie of God…Or if we admit of Tecla’s Edition, When I shewed unto you the mysterie of God, that is, revealed unto you the secret of the Gospel, or the hidden things of God’.

It is clear that Featley did not always see the need to decide between two variants, but there are occasions when he did. What were his criteria for doing so? Featley did not simply advocate for the majority reading. For example, he notes that in 1 Timothy 2:6, the word μαρτύριον (‘testimony’) ‘is left out in the Greek Copie written by Tecla, and the sense is full without it’. He goes on however: ‘if we retain the word because most copies have it, the meaning is…’ Being willing to consider a minority reading because it is contained in an older manuscript points to a belief that manuscripts should be weighed, rather than simply counted. With so many more manuscripts collated in our own day, we now know that Alexandrinus stands alone in its omission of μαρτύριον, and so no critical text has ever excluded it. Indeed, in the case of 1 Corinthians 10:3, where Featley (rightly) believed that Alexandrinus alone excluded the pronoun αυτο, he was not willing to accept the reading of a single manuscript, no matter how ancient. ‘Though the Pronoune αυτο be not in the addition [edition?] by Tecla, yet it appeareth by all other ancient Copies and the Commentaries of Augustine and Chrysostome upon this Text, that it ought to be added, and that our reading is the true’. In the one place where Featley explicitly rejects a Textus Receptus (and Majority Text) reading in favour of Alexandrinus, the reason is a combination of internal and external evidence. The occasion is 2 Timothy 4:14, where what is a curse in the TR (‘the Lord reward him’), is a prediction in the CT (‘the Lord will reward him’). The difference is the form of the verb: ἀποδῴη v. ἀποδώσει. Featley states: ‘This is not a curse, proceeding from anger, or a revengefull heart, which is contrary to the doctrine of Christ, Matth. 5.44. and of Saint Paul himself, Rom. 12.14. but a Propheticall commination out of a godly zeal, and a prediction inspired by the holy Ghost of the punishment which was ready to be executed upon him’. Featley therefore rejects a TR reading for doctrinal reasons, knowing that he has manuscript evidence for doing so: ‘with this interpretation agreeth the Greek Copie, of great Antiquity, written, as it is supposed by Tecla’.

How significant is all this?

It would be easy for a TR-onlyist to claim that all the foregoing data proves is that there were a diversity of views on Scripture at the Westminster Assembly, as there were on a number of other matters. Goodwin and Featley are, after all, just two Divines out of a much larger number.

A number of factors would militate against drawing that conclusion however. Firstly, far from holding some sort of ‘evolutionary’, nineteenth-century view of Scripture, Featley is actually commended by ‘Confessional Bibliologist’ Garnet Milne as having a higher view of the preservation of Scripture than either John Calvin or TR editor Theodore Beza! When it comes to the apparent contradiction between Genesis 33:19 and Acts 7:16, Featley says that Calvin’s solution ‘that there is an errour in all our copies of the New Testament, and ought to be corrected’ is ‘somewhat too peremptory’.[53] Featley also rejects Beza’s proposed conjectural emendation here, stating that ‘unlesse he could produce some ancient copies, wherein such mistakes were not to be found, openeth a dangerous gap to Infidents and Heretickes, who hereby will be apt to take occasion to question the infallible truth of the holy Writ’.[54]

Another thing to remember is that the access most Westminster Divines could have had to Codex Alexandrinus was very limited. The only access to all its readings was through physical access to this ‘jewel of antiquity’ kept in the Royal Library. (A number of continental theologians working on commentaries requested, and obtained, limited collations from the librarian. One of these was the Dutch scholar Hugo Grotius, and so the publication of his Annotationes in the 1640s became the occasion when ‘hundreds of readings from Codex Alexandrinus’ were ‘broadcast for the first time’.[55]) The contributors to the Westminster Annotations were ‘furnish’d with whatsoever books were needful’ by order of Parliament’s Committee for Religion. Other Divines were unlikely to have had the same access to Alexandrinus as Featley – as indicated by the growing desire by some of them in the early 1650s that Alexandrinus would be published. By the time collations from it were finally published in the London Polyglot in 1657, many of the members of the Assembly had died.[56] Thomas Goodwin was one of the last surviving members of the Assembly, and lived long enough after the publication of the Polyglot to be able to make use of it. Even after the publication of the Polyglot however, when fellow Westminster Divine Anthony Burgess appears to reference Alexandrinus, his only information comes from Grotius: ‘Grotius speaketh of a Manuscript that readeth…’.[57]

Acknowledging that most Westminster Divines could not have had access to Alexandrinus is one thing. It could still be claimed that the vast majority of Divines would have rejected non-TR readings if presented with them. For that to be the case however, it would need to be demonstrated that a majority of Divines believed that the Textus Receptus was jot and tittle perfect. As far as I am aware, the only attempt to do anything like that is Garnet Milne’s Has the Bible been kept pure?: the Westminster Confession of Faith and the providential preservation of Scripture. Milne attempts to go back to the sources to prove a ‘consensus view’ in favour of a TR-only position, which was only later ‘disrupted’.[58] However none of Milne’s many quotations actually provide evidence that any Westminster Divine would have rejected non-TR readings out of hand. Milne in fact explicitly claims Goodwin, Featley and Leigh – all of whom this article shows appealing to Alexandrinus over the TR – for his position! Milne also devotes a chapter to Ussher, who was enthusiastic about Alexandrinus and helped in the publication of both the Annotations and Walton’s Polyglot. Moreover, Milne’s footnotes not only include appeals to the Westminster Annotations (whose very title, as we have seen, advertises ‘various readings’ observed), but also to Matthew Poole. While not a Westminster Divine, Poole was one of the most trusted Reformed commentators of that or any other century – and frequently cited Alexandrinus (‘that most ancient manuscript’).[59] 

Even if it could be demonstrated that no Westminster Divine was a TR-onlyist, the final appeal of the ‘Confessional Bibliologist’ would be to the fact that the proof texts of the Westminster Confession contain verses which are in the Textus Receptus and not the Critical Text (or the Majority Text in the case of Acts 8:37 and 1 John 5:7 – which aren’t even in every edition of the Textus Receptus!). The Divines however specifically refused to build any doctrine on a text when they were aware that Greek evidence for it was lacking. The ‘Preface relating to the Author’ in the Works of Westminster Divine John Lightfoot recounts: ‘When some in the Assembly were for gathered churches, which must consist only of saints, and produced a place for that purpose taken out of the Revelations; chap. 15:3, “Thou King of saints” (Ἁγίων), Dr. Seaman [the second most frequent speaker at the Assembly and who served on 80 committees] well objected against it; because the reading was doubtful, some copies reading Αἰώνων instead of Ἁγίων, i.e. “Thou King of ages,” or “everlasting.” And Lightfoot backed him, by showing, that the Syrian and Arabic read to the same import’.[60] (Unlike those who framed the Confession, modern-day ‘Confessional Bibliologists’ will defend the ‘King of Saints’ reading to the hilt, even though it appears in no Greek manuscript prior to the publication of the Textus Receptus).[61]

Milne posits a consensus view in which the TR could not be questioned, but does not provide any evidence that such a view actually existed. If Goodwin and the Westminster Annotations were departing from the ‘consensus view’, surely we would expect their use of Alexandrinus to be controversial. There is no evidence that it was. Indeed, on the morning of 11 June 1641, ‘The authority of that most ancient Parchment MS. copy of the Bible remaining in his Majesties Library at Saint James, being all written in great Capitall Greeke Letters, was vouched and asserted’ in Parliament. The occasion was a bill about Bishops, and the appeal to Alexandrinus was to show that the postscripts included in the Textus Receptus at the end of the epistles to Timothy and Titus were ‘but the bold and spurious additions of some Easterne Bishop or Monke’.[62]

The writing was on the wall for the Textus Receptus. It has been said that even in the next century, ‘the Received Text was still treated with excessive veneration, and was not actually replaced in England until the nineteenth century. But events in the scholarly world had been gradually bringing about its decline, ever since the arrival of the Codex Alexandrinus (A) in 1627’.[63]

Some of the first to make use of Codex Alexandrinus were Westminster Divines. Richard Muller can describe an ‘older exegetical tradition’ including ‘such massive compilations as…the so-called Westminster Annotations’ and ‘the major works of Matthew Poole’.[64]

Right there, on page after page of that older exegetical tradition, are readings from Codex Alexandrinus.

About the author

Stephen Steele is minister of Stranraer RP Church in Scotland. He has an MA from Queen’s University Belfast where his focus was on nineteenth-century Presbyterianism.


[1]  Confessional Bibliologists, however, often fail to give a direct answer to the question of which of the many differing TR editions has been kept pure in all ages. See Mark Ward, ‘Which Textus Receptus? A critique of confessional bibliology’ in Detroit Baptist Seminary Journal, xxv (2020), 51-77.

[2]  Garnet Milne argues that ‘A fixed theology…requires a fixed text’, and so to accept a ‘previously unknown variant discovered at some future date’ which would change the meaning even “minutely”’ would ‘either be an addition or subtraction to the text, or an acceptance that the text up until that point had been deficient, corrupt and erroneous’. Garnet Howard Milne, Has the Bible been kept pure? The Westminster Confession of Faith and providential preservation of Scripture ([independently published], 2017), p. 145. Codexes Sinaiticus and Vaticanus are characterised as ‘Satan’s Bible’ in a recent work by ‘Confessional Bibliologists’: Jeffrey T. Riddle and Christian M. McShaffrey (eds),Why I Preach from the Received Text (Winter Springs, FL: 2022), p. 162.

[3]  Milne, Has the Bible been kept pure?, p. 301.

[4]  Gavin Beers, ‘From Atheism to the Authorized Version’ in Why I preach from the Received Text, p. 47.

[5]  Milne, Has the Bible been kept pure?, p. 135.

[6]  Chad Van Dixhoorn, The Minutes and Papers of the Westminster Assembly (Oxford,, 2012), i, 213; T. M. Lawrence, ‘Goodwin, Thomas (1600–1680)’ in Oxford Dictionary of National Biography (Oxford, 2004; online edn, Jan 2008).

[7]  John Lightfoot, The whole works of the Rev. John Lightfoot, ed. John Rogers Pitman (13 vols, London, 1822-5), i, xv.

[8]  David S. Katz, God’s Last Words: Reading the English Bible from the Reformation to Fundamentalism (New Haven and London, 2004), pp 88-90; John Eadie, The English Bible: An External and Critical History of the Various English Translations of Scripture, with Remarks on the Need of Revising the English New Testament (London, 1876), ii, 343-7.

[9]  Thomas Goodwin, The works of Thomas Goodwin D.D., sometime president of Magdalene College, Oxford (12 vols, Edinburgh, 1861-5), iv, 288.

[10]  Dirk van Miert, The emancipation of Biblical philology in the Dutch Republic, 1590-1670 (Oxford, 2018), p. 117.

[11]  J. H. Bowman, ‘The Codex Alexandrinus and the Alexandrian Greek Types’, in British Library Journal, xxiv, no. 2 (1998), p. 169.

[12]  Bruce Metzger and Bart Ehrman, The text of the New Testament: its transmission, corruption and restoration (4th edn, Oxford, 2005), p. 67.

[13]The harmony of the gospels: in the words of the authorized version: with an account of ancient manuscripts and of the various translations of the Holy Scriptures (London, 1863), p. 359

[14]  Goodwin, Works, i, 299.

[15]  Walton was assisted in this endeavour by Westminster Divine John Lightfoot, Archbishop James Ussher and others. Goodwin’s sermons on Ephesians were preached in the early 1640s (ODNB). The debate between Walton and John Own is well known, though the differences between them are sometimes overstated. Owen praised ‘the usefulness of the work’ and held it in ‘much esteem’. The most recent analysis of the debate concludes that ‘although they differed, they actually agreed on the core issues’. See Russel T. Fuller, ‘John Owen and the traditional Protestant view of the Hebrew Old Testament’ in Southern Baptist Journal of Theology, 20.4 (2016), pp 82-83.

[16]  Edward Leigh, Annotations upon all the New Testament philologicall and theologicall: wherein the emphasis and eleganice of the Greeke is observed, some imperfections in our translations are discovered, divers Jewish rites and customes tending to illustrate the text are mentioned, many antilogies and seeming contradictions reconciled, several darke and obscure places opened, sundry passages vindicated from the false glosses of papists and hereticks (London, 1650), p. 342.

[17]  Ussher, Works, xvi, 187. Ussher eventually received a copy of the readings and sent them to Henry Hammond, who used them for his Annotations. See Henry Hammond, A paraphrase and annotations on all the books of the New Testament (2nd edn, London, 1659), i.

[18]  Talbot Baines Reed, A history of the old English letter foundries, with notes, historical and biographical, on the rise and progress of English typography (London, 1887), pp 201-2; Elizabethanne Boran. ‘Young [Junius], Patrick (1584–1652), librarian and scholar’ in Oxford Dictionary of National Biography (Oxford, 2004; online edn, May 2012)

[19]  Goodwin, Works, ix, 370.

[20]  The KJV translation of ‘Grecians’ undoubtedly follows the TR reading, but for the purposes of comparison I have used the NKJV’s less unambiguous translation. Most commentators who accept the majority reading resolve the difficulty of the verse by understanding ‘Hellenists’ as a reference to Greeks rather than to Greek-speaking Jews.

[21]  Goodwin, Works, v, 475.

[22]  F. H. A. Scrivener, A Plain Introduction to the Criticism of the New Testament (London, 1894), ii, 371.

[23]  ‘this did always in former times in the reading of it breed some jar in my thoughts, as if the words had not been, at least, rightly and orderly placed’.

[24]  Goodwin, Works, ix, 218.

[25]  Johann Albrecht Bengel, Gnomon of the New Testament, ed. M. Ernest Bengel and J. C. F. Steudel, trans. James Bryce (Edinburgh, 1860), iv, 70, n. 2.

[26]  Charles Hodge, A Commentary on the Epistle to the Ephesians (New York, 1858), p. 74.

[27]  Goodwin, Works, i, 299-300.

[28]  Cited in Milne, Has the Bible been kept pure?, p. 167.

[29]  James Ussher, The Whole Works of the Most Rev. James Ussher, D.D., ed. by Charles Richard Elrington, 17 vols. (Dublin: Hodges, Smith, and Co., 1864), xv, 35. English translation taken from Elizabethanne Boran (ed.), The Correspondence of James Ussher, 1600-1656 (3 vols, Dublin, 2015), i, 35.

[30]  This and the next two paragraphs are based on Dean George Lampros, ‘A New Set of Spectacles: The “Assembly’s Annotations”, 1645-1657’ in Renaissance and Reformation, new series, xix, no. 4 (1995), pp 33-46.

[31]  Richard A. Muller, ‘Scripture and the Westminster Confession’ in Rowland A. Muller and Rowland S. Ward, Scripture and Worship (Phillipsburgh, NJ: 2007) p. 5.

[32]  Muller, ‘Scripture and the Westminster Confession’, p. 21.

[33]  The wording is taken from the annotation on Romans 12:11.

[34]  Philip D. W. Krey and Peter D. S. Krey (eds), Reformation Commentary on Scripture: Romans 9-16 (Downers Grove, IL: 2016), p. 140.

[35]  John Aubrey, cited in Reed, A history of the old English letter foundries, p. 203.

[36]  Brian Walton, The Considerator Considered: or, A brief view of certain considerations upon the Biblia Polyglotta, the Prolegomena and Appendix thereof (London, 1659), pp 140-1

[37]  Reed, A history of the old English letter foundries, p. 201.

[38]  Reed, A history of the old English letter foundries, p. 203.

[39]  Bowman, ‘The Codex Alexandrinus’, pp 170-2.

[40] While Featley’s status as a KJV translator is debated, the most recent assessment of his life concludes that ‘he was indeed a translator’. Greg A. Salazar, Calvinist conformity in post-reformation England: the theology and career of Daniel Featley (Oxford, 2022), p. 24.

[41]  Arnold Hunt, ‘Featley [Fairclough], Daniel (1582-1645)’ in Oxford Dictionary of National Biography (Oxford, 2004; online edn, Jan 2008).

[42]  J. V. Fesko, The theology of the Westminster Standards: historical context and theological insights (Crossway, 2014), p. 403.

[43]  William S. Barker, Puritan Profiles: 54 influential puritans at the time when the Westminster Confession of Faith was written (Fearn, 1996), pp 49-50.

[44]  Cited in Muller, ‘Scripture and the Westminster Confession’, p. 21.

[45]  Anthony à Wood, cited in Barker, Puritan Profiles, p. 50.

[46]  Muller, ‘Scripture and the Westminster Confession’, pp 19-20.

[47]  Muller, ‘Scripture and the Westminster Confession’, p. 21.

[48]  Pooyan Mehrshahi, ‘The Christian Bible Can Be Trusted’ in Why I Preach from the Received Text (Winter Springs, FL: 2022), p. 148.

[49]  Jeffrey T. Riddle, ‘Book Review: Mark Ward, Authorized: The use & misuse of the King James Bible’, in Bible League Quarterly, No. 479 (October-December 2019), p. 30; Milne, Has the Bible been kept pure?, p. 90.

[50]  Bruce Manning Metzger, A Textual Commentary on the Greek New Testament, (2nd ed., London, 1994), p. 463.

[51]  TR-advocates often try to claim that the Critical Text is ‘less orthodox’, but that argument can work both ways – see for example Mark Ward’s interview with Paul Himes, ‘Places Where the KJV Is “Less Orthodox” in Jude’: https://www.youtube.com/watch?v=LJBp8uufkB8.

[52]  See ‘Appendix E’ in The Cambridge Paragraph Bible: Of the Authorized English Version, ed. F. H. Scrivener (Cambridge, 1873)

[53]  Calvin says in his commentary: ‘it is manifest that there is a fault [mistake] in the word Abraham…Wherefore this place must be amended’.

[54]  Cited in Milne, Has the Bible been kept pure?, p. 140. For Beza’s solution, see Jan Krans, Beyond what is written: Erasmus and Beza as conjectural critics of the New Testament (Leiden and Boston, 2006) pp 324-5.

[55]  Miert, The emancipation of Biblical philology, pp 117-8, 135.

[56]  For the years of death of the Divines, see Chad Van Dixhoorn, ‘Members of the Westminster assembly and Scottish commissioners (1643–1652)’, in Oxford Dictionary of National Biography (Oxford, 2004; online edn, May 2007).

[57] Anthony Burgess, An expository comment, doctrinal, controversial and practical, upon the whole first chapter of the second epistle of St Paul to the Corinthians (London, 1661), p. 404.

[58]  Milne, Has the Bible been kept pure?, p. 301.

[59]  Citing Grotius in his Synopsis Criticorum on Revelation 4:3, as translated by the Matthew Poole Project: https://www.fromreformationtoreformation.com/post/poole-on-revelation-4-3-glory-of-the-covenant-god

[60]  Lightfoot, Works, i, 149-50.

[61]  See, for example, Jeff Riddle (http://www.jeffriddle.net/2016/03/word-magazine-50-review-james-white-on.html) and https://www.kjvtoday.com/saints-or-nations-or-ages-in-revelation-153. Cf Metzger, Textual Commentary, p. 679.

[62]  Sir Simonds D’Ewes, The Greeke postscripts of the epistles to Timothy and Titus cleared in Parliament. And an occasional speech touching the bill of acapitation, or poll-money (n. p., 1641).

[63]  B. F. Harris, ‘Richard Bentley and the text of the Greek New Testament’ in Evangelical Quarterly, xxxiv, no. 4 (1962), p. 214.

[64]  Richard A. Muller, Post-Reformation Reformed Dogmatics (4 vols, Grand Rapids, MI: 2003), iii, 135.