Charles Lee Irons, “A lexical defence of the Johannine ‘Only Begotten'” in Retrieving Eternal Generation. Edited by Fred Sanders and Scott R. Swain (Zondervan, 2017), 98-116.

In the course of writing a response and review of this book as a whole, I thought it worthwhile to respond to this chapter individually, as it is a chapter that I’ve thought a great deal about, having read some of Irons’ pre-published work on the topic and engaged with all of the same data.

Irons’ argument is that monogenes in the Johannine texts should be understood as “only-begotten,” not “only” or similar translations that have dominated NT scholarship for the past 60 years (largely thanks to Dale Moody’s 1953 article).

Irons does not provide any explanation of what “only-begotten” means. In my view, this is problematic precisely because “only-begotten” is not contemporary English and has no meaning for most English speakers outside a theological jargon phrase applied to the second person of the Trinity. That in itself suggests that “only-begotten” carries more semantic “freight” than we’d like. Connected to this, it suggests a significance or ‘weight’ upon the “begotten” part of that phrase that I’m not convinced is borne out by the data.

On page 99, Irons provides an occurrence count for μονογενής in fourth century fathers. This data affirms the prevalence of the term, but Irons offers no interrogation of those usages. A problem I will return to at his conclusion.

While Irons is generally correct that μονογενής is first translated into Latin as unicus filius (p100), his conclusion that this “suggests the -genēs stem was taken as communicating the notion of sonship or offspring” is too confident. It assumes that Latin translators attempted to divide μονογενής and render it by its constituents, which is unlikely

Irons’ fundamental argument that the historical development of a position that it mans “only” or “one and only” or “unique”, which he traces on p100 to 101, is accurate for that development. Similaraly, he is quite correct on p103 in arguing that Arians were more than happy to accept μονογενής differing only on the word’s interpretation. I believe that Irons’ attempt to clear away the etymological objection (p103-105) is well done, though anticipated mostly by Büchsel’s entry in the TDNT.

While Irons is correct that “only-begotten” does not develop in response to Arianism, his reliance upon Latin terminology on p102 is misleading, because unigenitus carries more genitus “weight” than μονογενής does.

Irons gets closest to what I believe to be the case when he writes, “the earliest meaning of monogenēs was biological, in reference to an only child.” (105) I puzzle why he did not stick with “only child”, which is the meaning I propose, but instead opts for the problematic “only begotten”. When he expands this to “monogenēs is used most basically and frequently in reference to an only child begotten by a parent, with the implication of not having any siblings,” the import of “begotten” seems to me problematic. “Only child” is all that is needed here, and the absence of siblings is the key factor in that. Every child is begotten, that’s a corollary of being a child, it doesn’t need to be imported into the definition. Indeed the definition is the absence of siblings, not the begottenness which is by necessity true of every child, only or otherwise.

Irons provides a raft of 4th century references in note 28, p106, some of which overlap my own recent SBL paper. I would suggest, though, that he at points confuses definitions from explanations, with some of these designating ‘what’ μονογενής means, others exploring and explaining the significance of the term.

Irons coverage of the classical occurrences of the term is comprehensive, and having covered all these texts myself, I can confirm the basic correctness of the claim that these use the word in a basic sense to refer to (in my terms) an only child.

Irons goes on to interpret the use in the case of Isaac, as a “nonliteral” extension. This is, going back to Carson’s chapter, the point Carson stumbled at because Isaac is not “literally” an “only-begotten” or an “only child”. Honestly, this is not really a stumbling block if you just have some common sense – anyone can call someone an only child, even if they aren’t, and by doing so they suggest a raft of implicatures. In the case of Isaac, he is the only child that counts, the child of the promise, or in Irons’ terms, the legitimate heir.

Irons sees the other uses of μονογενής as further metaphorical extensions (p110) I’m not so sure that they are extensions of the begetting concept, as it seems entirely possible that on the basis of the γένος etymology, they could arise as appropriate independent uses of μονογενής. But in any case, Irons is correct that when applied philosophically to the universe or similar, it designates it as unique. I do not think “only-begotten” works here as well as he does. Stronger, however, is the scientific use (p111) applied to trees and birds.

Returning them to John 1.14 and 1.18, the key texts, I think Irons’ argument is half-correct. For, the word “son” being absent many English translations must supply it, for “only son”. Irons thinks this is not far enough, and “only begotten son” is better. This, I feel, is part of his over-emphasis on “begottenness”, and part of the difference between meaning and translation. μονογενής on my argument means or designates “an only child”, but in attributive usages “child” is often specified with a noun – παῖς, θυγάτηρ, υἱός, in which case in translation you have to substitute that noun for “child”. But as a substantive, “only child” will work fine depending on context. Here, it is most natural to supply “son”. Adding “begotten” into “only son” to form “only begotten son” lays an unnatural emphasis on “begotten” in English.

Irons goes on to discuss Jn 1.18 and both NIV and ESV renderings. I concur that the ESV rendering gets it wrong, and to be honest Dale Moody is probably to blame. The NIV is slightly having a bet each way.

My final critique rests with Irons’ final conclusions. “The five occurrences of the term in the Gospel and first Epistle of John thus provide part of the exegetical basis for the traditional doctrine of the eternal begetting or generation of the Son, which is in turn a crucial linchpin for the pro-Nicene doctrine of the Trinity.” (116) Part, perhaps, but a small part. I refer you to my SBL presentation for the argument that these five occurrences do not directly provide an exegetical basis for the doctrine of eternal generation, and do not do so because they do not lay the emphasis on begetting that Irons does. He goes on, “the importance of the Johannine monogenēs for the construction of the doctrine of the eternal generation of the Son cannot be underestimated.” I presume he means “should not be underestimated”. In any case, again I must demur, because the reliance upon the Johannine monogenēs is not as great as suggested, and to return to the table on p99, this is misleading because actual reliance upon these Johannine texts is much, much less than Irons makes it out to be.

In conclusion, I am largely sympathetic to Irons’ argument that “only” or “unique” is a wrong understanding of the term μονογενής. Like Irons, I would point to the overwhelming body of Greek textual evidence that suggests that whenever it refers to a person, it designates a child without siblings (or intends a child with siblings to evoke the associations of an only child). I differ from Irons in three main, but important respects. Firstly, I think “only-begotten” ties us to a Latin trajectory that places weight upon the “begotten” part of that phrase more than that Greek term itself does. Secondly, that use of “begotten” reads more into the Johannine texts that is warranted of a doctrine of eternal generation per se. Thirdly, and the subject of my recent SBL paper, neither μονογενής as a term itself, nor the Johannine texts, provide as much direct grounding for pro-Nicene doctrines of eternal generation as Irons suggests.