When is an angelos not an angel? A critique of Christadelphian lexical semantics

There are a number of passages in the New Testament which appear, in standard English translations, to refer to bad angels. The most striking of these are Matt. 25:41, Rom. 8:38, 1 Cor. 6:3, 2 Pet. 2:4, Jude 6, and Rev. 12:7-9. Other possible references to bad angels in the New Testament include 1 Cor. 4:9, 11:10, 2 Cor. 12:7, 1 Pet. 3:19-22, and Rev. 9:11. Belief in bad angels was certainly prevalent in Second Temple Judaism, and was rooted in the interpretation of certain Old Testament passages (Gen. 6:1-4; Job 1-2; Dan. 10:13-21; see also Job 4:18; Psalm 78:49).

Now, Christadelphians hold that the Bible does not teach the existence of bad angels. How can such a claim be maintained in the face of the passages referred to above? The general approach is to deny that the word ἄγγελος )angelos)[1]  refers to angels in these texts. In some cases it may be allowed that the text refers to angels but denied that it refers to bad angels.

The purpose of this post is not to engage in a thorough exegesis of the above passages, but to offer some observations on the meaning of angelos in these texts. Correctly identifying the meaning of this word is crucial in Matt. 25:41, 2 Pet. 2:4, Jude 6 and Rev. 12:7-9, since in these four texts there is absolutely no question that the angeloi are morally bad (though strong arguments can be made for the moral badness of the angeloi in the other texts, too). In fact, in the case of 2 Pet. 2:4 and Jude 6, some Christadelphian exegetes now acknowledge that the text refers to bad angels, but proceed to claim that the writers were not actually asserting the existence of such beings (see a critique of this approach here and here).

In these four texts, nearly all English Bible translations[2] have translated angeloi with 'angels'. So too, all the major French translations[3] have 'anges', and all the major German translations[4] have 'Engel'. Thus, when Christadelphians claim that these angeloi are human beings rather than angels, they are contesting an overwhelming consensus. As far as I can tell, the only translation which does not render angeloi with 'angels' in these two texts is Young's Literal Translation. However, this is not very significant since the YLT always renders angelos with 'messenger'. While Young no doubt sought to bring out the etymological background of the word angelos, and while 'messenger' remained the primary meaning of angelos outside Judaism and Christianity, the usual meaning of angelos in the New Testament is 'angel'.

Martin argues that even within the Hebrew Bible, the word מַלְאָךְ (mal’āk) gradually became a technical term rather than a common noun; and when the Hebrew Bible was translated into Greek (as the LXX), "the Jewish scholars translated the term מַלְאָךְ into ἄγγελος and thus introduced a new technical term, one referring to a particular species, into Greek."[5] Martin explains what he means by distinguishing a technical term from a common noun:

I simply mean the difference between a noun that refers to a recognized class of beings rather than to an activity or role. 'Golden Retriever' does not refer merely to a yellow dog that retrieves.[6]

Similarly, discussing semantic change in the New Testament, Silva comments on the frequent reduction in the meaning of words; words become 'specialized' relative to their previous usage and their usage in non-Christian literature. He explains the exegetical significance of this phenomenon:

We must understand that once the semantic range of a term has been narrowed, we are less dependent on the context when we wish to grasp the meaning of the word. That is, the word becomes more precise: a more or less definite referent (what the word stands for) is automatically associated with the word itself. These are the terms that become technically charged at times, so that they serve as ‘shorthand’ for considerable theological reflection.[7]

Silva includes angelos in a list of "(more or less technical) specializations".[8] Other examples for comparison are ἐκκλησία (ekklēsia), which was specialized from 'assembly' to 'church' (ecclesia if you prefer), and εὐαγγέλιον (euangelion), specialized from 'good news' to 'gospel (message of salvation)'.

An interesting point is made by Turner on the emergence of the term ἀπόστολος (apostolos). This term is never used in the LXX, whereas angelos is occasionally used for a human who brings the divine message (2 Chr. 36:15-16; Isa. 44:26; Jer. 29:14(49:14)[9]; Hag. 1:13; Mal. 1:1; 2:7; 3:1). (Indeed, over 40% of instances of angelos in the LXX refer to human messengers).[10] Turner explains the emergence of the term apostolos in the New Testament on the grounds that "the preachers rejected angelos (messenger) as already a technical term for 'angel'".[11] Thus in the New Testament we see a conscious shift away from the use of angelos for humans.

The emergence of angelos as a technical term means that the mere use of the word angelos is generally sufficient to establish the meaning 'angel', apart from contextual considerations. We can see this in a number of New Testament passages in which angelos is used with no explicit reference to sending and no qualifier such as ‘of the Lord’ or ‘in heaven’: Matt. 4:11; 25:31; Mark 1:13; Luke 16:22; 20:36; 24:23; John 12:29; Acts 6:15; 7:30; 23:8; Rom. 8:38; 1 Cor. 4:9; 6:3; 11:10; 13:10; Gal. 3:19; Col. 2:18; 1 Tim. 3:16; Heb. 1:4; 2:5; 2:7; 2:9; 2:16; 12:22; 13:2; 1 Pet. 1:12; 3:22; 2 Pet. 2:11; Rev. 5:11; 7:11; 12:7; 21:17.

To quote a few of these examples:

Then the devil left him, and behold, angels came and were ministering to him. (Matt. 4:11 ESV)

That is why a wife ought to have a symbol of authority on her head, because of the angels. (1 Cor. 11:10 ESV) 

Let no one disqualify you, insisting on asceticism and worship of angels, going on in detail about visions, puffed up without reason by his sensuous mind, (Col. 2:18 ESV)

It was revealed to them that they were serving not themselves but you, in the things that have now been announced to you through those who preached the good news to you by the Holy Spirit sent from heaven, things into which angels long to look. (1 Pet. 1:12 ESV)

In all of these passages, the mere use of the word angeloi is sufficient to impart the meaning. No contextual clues are given, and none are needed. This shows why angelos is said to be a technical term in its Christian usage: one does not first read 'messenger' and then reflect on the concept of 'messengers' and ask, 'What sort of messenger is this?' One simply reads 'angel', and rightly so. In the same way, if your friend tells you he got a golden retriever, you do not reflect on the concept of 'retrieval' to get his meaning; you simply identify this as a technical term for a particular breed of dog.

If a New Testament writer wanted to use this word as a common noun meaning 'messenger' with a human referent, he would need to make this clear from the context to ensure his Christian readers did not apply the usual technical meaning. Indeed, this is exactly what we find in the six New Testament texts where angelos is generally regarded as referring to human messengers.

Three of these are direct quotations from Mal. 3:1 LXX (in Matt. 11:10; Mark 1:2; Luke 7:27). In quoting Scripture a writer would be less likely to introduce his preferred vocabulary than in his own original material. Moreover, in Mal. 3:1 LXX and in all three of the Gospel quotations, the verb ἀποστέλλω (apostellō, ‘send’) is used, which highlights the functional meaning of angelos.[12] Moreover, the application of the prophecy to John the Baptist (implicitly in Mark but explicitly in Matthew and Luke) makes the human referent of angelos unmistakable.

Two further non-technical uses of angelos are in the Gospel of Luke (7:24; 9:52). Being probably the sole Gentile among the New Testament writers, Luke may have been more accustomed to using angelos with its secular, non-technical meaning. In any case, in these two texts, as in his quotation from Malachi, he makes both the functional meaning of angelos and the human referent perfectly clear. In Luke 7:24, John's angeloi are those disciples who have been sent (apostellō) to Jesus as envoys with a message (Luke 7:19-20). In Luke 9:52, the angeloi are those disciples whom Jesus sent (apostellō) ahead of him to a Samaritan village.

A final non-technical use of angelos is in James 2:25, where the word refers to the Israelite spies received by Rahab and sent out another way. Here too, the envoy function of the word is apparent. Moreover, although angelos does not occur in the LXX account of Rahab and the spies, mal’āk is used twice of the spies in the Hebrew text (Josh. 6:17; 6:25). This fact is sufficient to account for James' decision to use angelos in a non-technical sense here.

The two other cases in the New Testament in which angelos may be used non-technically are the reference to the angelos satana in 2 Cor. 12:7 and the references to the angeloi of the seven churches in Rev. 1:20; 2:1; 2:8; 2:12; 2:18; 3:1; 3:7; 3:14. In the former case, English translations almost unanimously render angelos as 'messenger', implying that 'messenger of Satan' is a metaphorical description of Paul's thorn in the flesh. The tendency among English translations (which probably owes much to the precedent set by the KJV) is not found in other languages. For instance, three out of four major French translations (NEG1979, LSG, SG21) have "ange", while all five major German translations (Luther, HOF, NGU-DE, SCH1951, SCH2000) have "Engel". Furthermore, there has been an emphatic shift in recent English-language scholarship toward interpreting the angelos of 2 Cor. 12:7 as an angel.[13] In any case, it is virtually certain that the angelos of 2 Cor. 12:7 is not a human person.

As for the angeloi of the seven churches, translations are nearly unanimous in rendering the word with 'angels'. The precise sense is obscure; Moulton explained it in terms of the 'representative angels' concept which he also sees behind texts such as Matt. 18:10 and Acts 12:15.[14] Hemer mentions five possible interpretations:

A choice is generally offered between (1) heavenly guardians of the churches, and (2) human representatives of them, generally their bishops. Three other principal variants deserve consideration: (3) that the 'angels' are personifications of the churches; (4) that they are literally human 'messengers'; and (5) that the term is used in some complex and elusive way or at differing levels, so that we cannot expect to assign it a lexical equivalent that tells the whole story.[15]

He adds that "Of the theories proposed we may most easily criticize (2) and (4)", which he proceeds to do. Osborne too mentions five views, and while he notes that the solution to this exegetical problem is not simple, ""the use of 'angel' in this book makes it extremely unlikely that these are human 'messengers' of any type".[16] Mounce likewise argues that "The use of 'angel' in the book of Revelation (it occurs some 60 times) favors identifying the angels as heavenly beings".[17] Johnson adds,

A strong objection to the human messenger sense here is the fact that the word is not used this way anywhere else in apocalyptic literature. Furthermore, in early noncanonical Christian literature no historical person connected with the church is ever called an angelos.[18]

Once again, then, despite the exegetical difficulties here, there is little support for interpreting the angeloi of the seven churches as human beings.

Thus, aside from the 'bad angeloi' texts under consideration, there are only six instances in the NT in which angelos refers to a human being. In four of these six instances, the usage can be explained by quotation from or dependence on an OT passage in which angelos is used in the LXX or mal’āk in the Hebrew. In the two remaining cases, the functional, non-technical use of angelos is made unmistakably clear from the context, which uses the verb apostellō and explicitly identifies the angeloi as human persons.

Thus, in all six generally accepted cases of angelos referring to a human being in the NT, the envoy function is made explicit in the context, making obvious the reason why the word angelos was chosen. Additionally, in each of these cases a human referent is specified by the writer: John the Baptist, or John’s disciples, or Jesus’ disciples, or the spies who visited Rahab.

This brings us back to four texts which undoubtedly refer to 'bad' angeloi but which Christadelphians typically interpret to refer to bad human beings: Matt. 25:41, 2 Pet. 2:4, Jude 6, and Rev. 12:7-9. The question is whether angelos carries its usual technical meaning, or whether there are contextual grounds for interpreting it non-technically. First of all, none of these four texts quote from or allude to an OT text in which mal’āk or angelos occurs. Secondly, none of these four texts contain any functional language pertaining to sending or a message which might direct us toward the non-technical meaning 'messenger'. Thirdly, none of these four texts explicitly identify the referents of angeloi as human beings.[19]

We should also note that there is absolutely no justification for broadening the semantic field of angelos as in Heaster’s claim that angelos means "a messenger or, by extension, a follower".[20] The BDAG lexicon attests two primary meanings for angelos: "a human messenger serving as an envoy; an envoy, one who is sent" and "a transcendent power who carries out various missions or tasks; messenger, angel".[21] The LSJ lexicon (which covers the whole of antiquity and does not restrict its interest to Christian literature), gives four meanings: "messenger, envoy; generally, one that announces or tells; angel; in later philosophy, semi-divine being".[22] In other words, although one can understand why a Christadelphian exegete would want to introduce a more general term such as 'follower', it simply does not fall within the semantic range of the word angelos.[23]

Thus, in complete contrast to the six cases considered above, there is no positive evidence for regarding Matt. 25:41, 2 Pet. 2:4, Jude 6 or Rev. 12:7-9 as exceptions to the usual technical NT meaning of angelos, namely, ‘angel’. Moreover, parallels in Second Temple Judaism to all four of these texts provide positive evidence for understanding the angeloi in these texts to be angels (the Jewish background to Matt. 25:41 will be discussed in a future post). In the case of Rev. 12:7-9, the angeloi of the dragon (which is a symbolic reference to the devil, as v. 9 makes clear) are pitted in a war against Michael and his angeloi. One cannot escape the obvious implications simply by drawing attention to symbolic elements in the wider context. Two group of angeloi are doing battle. One group is unquestionably angelic. How can we possibly reach the conclusion that the other group is not? (Note Johnson's statement above that angelos is not used of human messengers anywhere else in apocalyptic literature).

We are now in a position to answer the titular question. When is an angelos not an angel? In the New Testament, the Greek noun angelos (meaning a messenger or envoy) has become a technical theological term meaning 'angel'. There are six exceptional texts where the non-technical meaning of ‘messenger’ or ‘envoy’ is applied to human beings, but these can be identified using three simple criteria: (1) the presence of functional 'sending' language; (2) explicit identification of the angelos as a human being; and (3) quotation from or allusion to an OT passage in which angelos or mal’āk is used in this way. Criteria (1) and (2) are met in all six NT 'human messenger' texts, and criterion (3) is met in four out of six. By contrast, none of these three criteria are met in any of the four 'bad' angel texts we have discussed (or, for that matter, the other NT texts mentioned in the first paragraph).

We can safely draw the conclusion that has been virtually uncontested in biblical scholarship from the patristic era up until the present day: the New Testament teaches the existence of bad angels.

---

[1] The Greek word ἄγγελος is sometimes transliterated as aggelos, but more frequently as angelos, since a double γ made a ‘ng’ sound in ancient Greek.

[2] KJV; NKJV; RSV; NRSV; ASV; NASB; ESV; ISV; NIV; NET; NLT; etc.

[3] BDS, LSG, NEG1979; SG21.

[4] Luther; HOF; NGU-DE; SCH1951; SCH2000.

[5] Martin, D.B. (2010). When Did Angels Become Demons? Journal of Biblical Literature, 129(4), p. 665.

[6] Martin, D.B. op. cit., p. 664 n. 34.

[7] Silva, M. (1994). Biblical Words and their Meaning: An Introduction to Lexical Semantics (Revised and Expanded Edition). Grand Rapids: Zondervan, p. 77. Emphasis added.

[8] Silva, M. op. cit., p. 79.

[9] The confusing reference to this text is due to the disparity between the Hebrew and LXX text of Jeremiah. It is found at Jer. 29:14 in Brenton’s LXX translation and the NETS at 30:8 in other sources. It corresponds to Jer. 49:14 in the English Bible.

[10] This statistic is based on my own analysis of in ἄγγελος the LXX but follows the NETS translation.

[11] Turner, N. (1981). Christian Words. Nashville: Thomas Nelson, p. 25.

[12] In most cases in the LXX where angelos is used with a human referent, the context explicitly contains the idea of sending, and often the content of the envoy's message: see e.g. Gen. 32:4-7; Num. 20:14; Num. 24:12; Josh. 7:22; Judg. 6:35; 7:24; 9:31; 11:12-19; 1 Sam. 6:21; 11:3-9; 16:19; 19:11-21; 23:27; 25:14; 2 Sam. 2:5; 3:12-14; 3:26; 5:11; 11:4-25; 12:27; 1 Kings 20:5-9; 22:13; 2 Kings 1:2-16; 5:10; 6:32-33; 2 Kings 7:15-17; 9:18; 10:8; 14:8; 16:7; 17:4; 18:14; 19:9; 19:14; 1 Chr. 19:2; 19:16; 2 Chr. 18:12; 35:21; 36:15-16; Neh. 6:3; Job 1:14-18; Isa. 18:2; 33:7; 37:9; 37:14; Jer. 27:3; Ezek. 17:15; 23:16; 23:40; Hag. 1:13; Mal. 2:7.

[13] Price, R.M. (1980). Punished in Paradise: An Exegetical Theory on II Corinthians 12:1-10. Journal for the Study of the New Testament, 7, pp. 33-40; Thomas, J.C. (1996). ‘An angel from Satan’: Paul’s thorn in the flesh (2 Corinthians 12.7-10). Journal of Pentecostal Theology, 9, pp. 39-52. Thrall, M.E. (2000). A Critical and Exegetical Commentary on the Second Epistle to the Corinthians (Vol. 2). London: Bloomsbury Academic, p. 808f; Williams, G. (2009). The Spirit World in the Letters of Paul the Apostle: A Critical Examination of the Role of Spiritual Beings in the Authentic Pauline Epistles. Göttingen: Vandenhoeck & Ruprecht, pp. 105-109; Martin, D.B. op. cit., p. 674; Wallace, J.B. (2011). Snatched into Paradise (2 Cor 12:1-10): Paul's Heavenly Journey in the Context of Early Christian Experience. Berlin: Walter de Gruyter, pp. 272-273; Becker, M. (2013). Paul and the Evil One. In E. Koskenniemi & I. Fröhlich (Eds.), Evil and the Devil (pp. 127-141). London: Bloomsbury T&T Clark, p. 136.

[14] Moulton, J. H. (1902). ‘It is his angel’. Journal of Theological Studies, 12, pp. 514-527.

[15] Hemer, C.J. (1986). The Letters to the Seven Churches of Asia in Their Local Setting. London: A&C Black, p. 32.

[16] Osborne, G.R. (2002). Revelation. Grand Rapids: Baker Academic, p. 99.

[17] Mounce, R.H. (1998). The Book of Revelation. Grand Rapids: Eerdmans, p. 63.

[18] Johnson, A.F. (2006). Revelation. In T. Longman III & D.E. Garland (Eds.), The Expositor’s Bible Commentary (Vol. 13) (pp. 571-789). Grand Rapids: Zondervan.

[19] Note that in Matt. 25:41, the fact that wicked humans are consigned to eternal fire prepared for the diabolos and his angeloi in no way implies that the wicked humans are the diabolos and his angeloi.

[20] Heaster, D. (2012). The Real Devil (3rd ed.). South Croydon: Carelinks Publishing, p. 409.

[21] Arndt, W., Danker, F.W., & Bauer, W. (2000). A Greek-English lexicon of the New Testament

and other early Christian literature (3rd ed.). Chicago: University of Chicago Press, p. 8.

[22] Liddell, H.G. & Scott, R. (1940). ‘ἄγγελος’. A Greek-English Lexicon (Revised and augmented throughout by Sir Henry Stuart Jones with the assistance of. Roderick McKenzie). Oxford: Clarendon Press.

[23] Note that in Isa. 37:24 LXX, angelos is used where the Hebrew Bible has עֶבֶד (‘ebed), the usual Hebrew word for 'servant'. However, this is because the angeloi referred to are in fact messengers. They have been referred to using the word mal’āk twice previously in the passage (Isa. 37:9; 37:14), and even in v. 24 the emphasis is on the words which they have brought against the Lord on Sennacherib's behalf.