The classical doctrine of the Trinity affirms that within the one, undivided being of God there are three persons—Father, Son, and Holy Spirit. It also affirms that what distinguishes the three persons are their relations of origin: the Father is unbegotten, the Son is eternally begotten of the Father, and the Spirit proceeds from the Father and the Son. The second point is referred to as the doctrine of the eternal generation of the Son. This doctrine has traditionally been grounded in a number of scriptural proof texts, one set of which is the five Johannine verses that, according to the Vulgate and the King James Version, affirm the Son is the “only begotten” Son of God (John 1:14, 18; 3:16, 18; 1 John 4:9).
In recent times, however, many evangelical theologians have doubted whether the doctrine of the eternal generation of the Son is indeed taught in Scripture. A principal source of doubt has been the 20th-century scholarly consensus that the Greek word monogenēs does not mean “only begotten.” Scholars have argued that the compound Greek adjective is not derived from monos (“only”) + gennao (“beget”) but from monos (“only”) + genos (“kind”). Thus, they argue, the term shouldn’t be translated “only begotten” but “only one of his kind” or “unique.” Reflecting the scholarly consensus, most modern English versions have adopted this new understanding and translate the five Johannine uses of monogenēs as “only” (CEV, ESV, NAB, NRSV, RSV) or “one and only” (HCSB/CSB, NIV, NLT). Only a few retain “only begotten” (NKJV, MEV, NASB).
This shift in the scholarly understanding of the term effectively removed a crucial scriptural underpinning for the doctrine that the Son is begotten of the Father. In theory, other proof texts could still be appealed to, but once this brick was removed, for some it seemed the whole wall was ready to fall.
In this article, I would like to offer a brief defense of the traditional translation “only begotten.”
Examining the Linguistic Data
First, a search of Thesaurus Linguae Graecae—a comprehensive database of ancient, Koine, and medieval Greek—reveals that the word monogenēs is used most basically and frequently in contexts having to do with biological offspring. Its fundamental meaning is “only begotten” or “only child” in the sense of having no siblings. For example, Plato describes the primitive population of the mythical island of Atlantis as follows:
Thereon dwelt one of the natives originally sprung from the earth, Evenor by name, with his wife Leucippe; and they had for offspring an only begotten daughter, Cleito. (Critias 113d; LCL)
This biological usage can be seen in the instances of monogenēs in non-Christological contexts in the New Testament. Three times, Luke uses monogenēs to describe various “only begottens” whom Jesus healed: “the only son” of the widow of Nain (Luke 7:12); the “only daughter” of Jairus (Luke 8:42); and the demon-oppressed boy whose father pleaded, “Teacher, I beg you to look at my son, for he is my only child” (Luke 9:38 ESV).
It must be acknowledged that there are indeed instances where a translation such as “only,” “only one of its kind,” or “unique” is required by the context. For example, Clement calls the phoenix, a creature he thought really existed, “unique” (1 Clement 25:1). An ancient treatise describes trees that exist in “only one kind.” But these are uniformly metaphorical extensions of the basic meaning, “only begotten” or “only child.” Context determines which usage is in view, and the five Johannine uses are in the context of sonship, not botany.
Second, careful examination of the word list of Thesaurus Linguae Graecae reveals at least 145 other words based on the –genēs stem. Three examples will suffice: theogenēs (“born of God”), neogenēs (“newborn, newly produced”), and, my personal favorite, konchogenēs (“born from a shell”; picture the birth of Venus). Of these 145 words, fewer than a dozen have meanings involving the notion of genus or kind—for example, homogenēs (“of the same genus”) and heterogenēs (“of different kind”).
In addition, there are at least 58 Greek proper names built on the –genēs stem, like the common Diogenēs (“born of Zeus”). Since these are names presumably given by parents to their children, we may assume they generally have some connection with the embodied reality of biological offspring, rather than the abstract notion of species or kind. The list of –genēs words and proper names continues to grow as we move forward into medieval (Byzantine) Greek. Taken together, this wealth of –genēs words constitutes critical data demonstrating that the –genēs stem strongly encodes notions of derivation, offspring, and begetting throughout the history of the Greek language.
But what about the etymological argument that the –genēs portion of monogenēs comes from genos (“kind”) rather than gennao (“beget”)? This argument collapses once it is recognized that both genos and gennao derive from a common Indo-European root, ǵenh (“beget, arise”). This root produces a fair number of Greek words having to do with biological concepts of begetting, birth, and offspring. In fact, the word genos itself sometimes means “descendant” (Rev. 22:16). True, it can also mean “kind” in a scientific or classification sense where literal biological descent is not in view (e.g., “different kinds of languages” [1 Cor 14:10]). But the scientific or classification usage is a metaphorical extension of the literal biological sense, since the abstract concept of “kind” is modeled on the embodied biological experience of the similarities shared by offspring descended from a common parent.
One common objection to the traditional translation has been the use of monogenēs in Hebrews 11:17:
By faith Abraham, when he was tested, offered up Isaac, and he who had received the promises was in the act of offering up his only son. (ESV)
Isaac wasn’t the only son of Abraham, since he had an older brother, Ishmael, from a different mother (Hagar). Therefore, the argument goes, the term doesn’t mean “only begotten” son but “unique” son.
But this objection fails to reckon with the inherent flexibility of language. It may not be literally true that Isaac is Abraham’s only son, but he can still be called “only begotten” to highlight the fact that he is Abraham’s sole heir. Ishmael has been rejected from the line of promise. Sarah told Abraham to cast out Hagar and her son, “for the son of this slave woman shall not be heir with my son Isaac” (Gen. 21:10). God agreed with Sarah and told Abraham to do as she said, “for through Isaac shall your offspring be named” (v. 12; quoted in Rom. 9:7; Heb. 11:18). As a result, it is “as if” Isaac is Abraham’s only begotten son. This “as if” usage of monogenēs is attested elsewhere in Greek literature.
Restoring the Brick
I continue to research the relevant Greek data, but it should be clear that a decent case can be made for rendering monogenēs as “only begotten” in the five Christological occurrences in the writings of John. To be sure, we mustn’t think the entire doctrine of the eternal generation of the Son stands or falls on this one word. After all, the word itself is silent on the question of eternal generation or begetting.
A number of other key proof texts and broader biblical-theological themes need to be brought to bear in order to fashion a robust case for the doctrine. Yet my research suggests we have good reason to restore one of the bricks in the wall of scriptural support for the belief that the Son is begotten of the Father, as the church fathers taught and as the church confesses in the Nicene Creed.
Author’s note: For a more detailed argument, see my contribution to the forthcoming book, Retrieving Eternal Generation, edited by Fred Sanders and Scott Swain (2017).
 Dale Moody, “God’s Only Son: The Translation of John 3:16 in the Revised Standard Version,” Journal of Biblical Literature 72 (1953): 213–19; Richard N. Longenecker, “The One and Only Son,” in The NIV: The Making of a Contemporary Translation, ed. Kenneth L. Barker (Grand Rapids: Zondervan, 1986), 119–26; Gerard Pendrick, “MONOGENES,” New Testament Studies 41 (1995): 587–600.
 This massive searchable repository operated by the University of California, Irvine, contains most Greek literary texts from the 8th/7th century BC to the fall of Constantinople in AD 1453.
 Lexicon of Greek Personal Names, vols. 1–5, ed. Peter M. Fraser et al. (Oxford: Oxford University Press, 1987–2013).
 Robert Beekes, Etymological Dictionary of Greek (Leiden: Brill, 2010), 1.266, 272–73.