Carl W.Conrad cwconrad at artsci.wustl.edu
Mon Oct 22 19:45:38 EDT 2007

On Oct 20, 2007, at 11:34 AM, Iver Larsen wrote:

>> From Iver:
> It looks like I need to clarify my position which is based on an  
> empirical analysis of all the NT
> data plus a semantic theory of case frames.

Iver, I do very much appreciate the pains you've been to to respond  
so completely to my request for clarification. As a result, I now  
have a much clearer sense of the points upon which we are in  
agreement as well as the issues upon which we have reached different  
understandings. I even understand what you mean by 'case-frame  
theory' -- I've worked through the work of Paul Danove (_Linguistics  
and Exegesis in the Gospel of Mark_) and found it generally helpful,  
although some instances of analysis of middle and passive forms has  
seemed to me questionable. I've a lot to say in response, but I've  
decided to send on some of it now and to deal with the rest of it in  
a second part focusing on the lexical problems; I hope to get that  
off tomorrow.

> ----- Original Message -----
> From: "Carl W. Conrad" <cwconrad at artsci.wustl.edu>
> To: "Iver Larsen" <iver_larsen at sil.org>
> Cc: "BG" <b-greek at lists.ibiblio.org>
> Sent: 19. oktober 2007 18:05
> Subject: Re: [B-Greek] BAPTIZW/BAPTIZOMAI
> CC:
> (1) Are we agreed that the -QH- aorists and futures express both
> middle and passive semantics, e.g. that APEKRIQH is identical in
> sense to APEKRINATO and is the more common later Hellenistic form of
> the aorist of APOKRINOMAI? I ask this because I rather think that
> there are many instances of the BAPTISQH/E- stem -- even within the
> GNT -- that are semantically middle and that were becoming or had
> become standard replacements for the older middles in BAPTISA-.  I've
> argued that specifically for those that have an accusative object (TO
> adduced another example from Pap.Oxy. 840, lines 9-16:
> Iver: We are agreed that the -QH- forms which are sometimes called  
> passive morphology and the middle
> morphology are at times skewed in relation to the semantics,  
> whether the sense is middle or passive.
> However, the situation varies with the particular word in question.  
> For some verbs like APEKRIQH,
> the -QH- forms have become common with an active sense, and there  
> is no apparent difference between
> an aorist middle and aorist passive of this verb. In the NT, 216  
> out of 231 (94%) use the QH. There
> are others in this category, e.g. DUNAMAI.
> For other verbs, there is a clear distinction between active,  
> middle and passive.
> Some verbs do not occur in the middle, only active and passive,  
> others do not occur in the active,
> others not in the passive.
> So, we need to look at the particular word in question and its  
> usage in context. What applies to one
> verb does not necessarily apply to another.

It appears that we are more fully in agreement here than I was  
thinking, although I find some of your phraseology problematic,  
especially when you speak of -QH- forms with an "active" sense.  
Perhaps this is something that requires further clarification; I may  
also be misunderstanding Danker's phraseology when he describes (in  
BDAG) some verbs as "passive with active sense." I have supposed that  
he is referring to the traditional notion of deponency, but he may be  
using the term "active sense" to indicate that the verb is transitive  
and takes a direct object; inasmuch as middle-voice forms, although  
often intransitive, are nevertheless commonly enough transitive and  
take a direct object, I find that use of "active sense" misleading.  
The morphology of a Greek verb doesn't necessarily have anything to  
do with transitivity; it might be better to speak of "agentive" verbs  
where the subject is the initiator of the action. I do think it's  
useful to distinguish between "agentive" and "non-agentive" middle- 
voice verbs.

I also think it is a little bit misleading to use the word "skewed"  
to describe the relationship between semantics and the middle and  
passive morphoparadigms (in the GNT, especially, which is what we're  
talking about primarily) -- but I think that we are in substantial  
agreement over the facts about the -QH- verb forms in Hellenistic  
Greek. My figures for APOKRINOMAI in the GNT aorist middle (7) and  
aorist 'passive' (213) are somewhat at odds with yours (perhaps  
you're counting MS variants), but the percentages are roughly  
identical. This has to do with what I like to refer to as NT Koine  
being a "language in flux" wherein earlier forms and usages are found  
in the same corpus, sometimes in the same authors even as later forms  
and usages. Clearly APEKRIQHN is the standard aorist form of the  
verb. On the other hand EGENOMHN is still the standard NT Koine  
aorist of GINOMAI while EGENHQHN is relatively rare. A less-common  
verb is AGALLIAOMAI: BDAG lemmatizes this as AGALLIAW but the middle  
is clearly the standard form, even if Lk 1:47 has an active  
HGALLIASEN, Rev 19:7 AGALLIWMEN and Jn 5:35 has an aorist infinitive  
AGALLIAQHNAI (the LXX has only one active form -- in the Odes of  
Solomon! -- in 72 instances.

I agree that we need, when analyzing the relationship of voice  
morphology to usage, to consider each verb on its own terms. I think  
that transitivity is one key factor here, and I also think that  
volition of the subject is a factor (as you seem not to think). I  
agree that some verbs are fundamentally transitive and are best  
described in terms of the two arguments "subject" and "patient." I  
think we may differ over the analysis of the usage of some verbs, and  
BAPTIZW/BAPTIZOMAI may be one of them.

> Randall did dig up a clear example of a passive morphology BAPTIZW  
> being used in a middle sense from
> this papyrus, but there is nothing like it in the NT, and it did  
> not have the sense of "baptism".
> Mk 10:38 I'll discuss later.
> CC:
> (2) Unless I have misunderstood, Iver holds that BAPTIZW in the
> cultic sense of "baptize" is strictly transitive and is used strictly
> of an action performed by one person upon a second person or persons.
> A direct reflexive usage of BAPTIZW in this sense, even with the
> sense, "have oneself baptized (by another)" is not thinkable or not
> attested.
> Iver: Yes, I believe it is not thinkable, nor attested. In terms of  
> case frame and semantics, if you
> have yourself baptized by another, it is not reflexive. It is only  
> reflexive if you do it to
> yourself, or in semantic terms, if the agent and patient is the same.
> CC:
> Instances that we do find of middle usage of this verb in
> the sense "baptize" are metaphorical (Mk 10:38) or to be explained
> otherwise, as Acts 18:8:  KAI NUN TI MELLEIS/ ANASTAS BAPTISAI
> Iver seems to think that the middle imperative is influenced by the
> other adjacent middles APOLOUSAI and EPIKALESAMENOS. My own view is
> rather that many of the 12 instances of middle forms of BAPTIZW/
> BAPTIZOMAI are semantically middle and that they do fall into the
> direct reflexive category in the sense "have oneself baptized" -- and
> that this direct-reflexive usage may be in play even when there's a
> hUPO + genitive construction indicating an external agent. How the
> academic linguists choose to explain a distinction between the agent
> of a direct reflexive of this sort and an external agent, I don't
> know, but I think that such an expression as "have one's hair cut by
> one's favorite barber" is perfectly intelligible and that it could be
> expressed unambiguously in ancient Greek as KEIRESQAI hUPO TOU
> Iver: Maybe what you are suggesting is that the subject could  
> simultaneously fill the role of
> patient and beneficiary? I have no quarrel with that, and one of  
> the many usages of the middle is to
> indicate that the subject is also the benificiary.  However, it is  
> not reflexive, because the agent
> is the barber, not the person whose hair is being cut.
> Kimmo recently mentioned that James 4:3 has an interesting contrast  
> between the active AITEITE and
> the middle AITEISQE. In the second, middle form there is focus on  
> the fact that the agent is also
> the benificiary. Both forms can have up to two objects: The thing  
> asked for and the person from whom
> it is asked. In semantic terms the patient and the source. The  
> source can for this verb be expressed
> either as an accusative or with a preposition.
> In James 4:2 we find the word MACESQE (fight). This word does not  
> have an active morphological
> paradigm, because it is inherently reciprocal and therefore middle.  
> You only fight if you have
> someone to fight with. It is normally, but not exclusively, used  
> with a plural subject. In Danish,
> we indicate such reciprocity by a final -s, e.g. "Han sk¾ndte pŒ  
> ham" (He rebuked (on) him), but "De
> sk¾ndtes" (They rebuked one another, or they had an argument). "Han  
> sk¾ndtes med hende" (He had an
> argument with her).
> So, the middle in Greek has several distinct functions, e.g.  
> reflexivity, reciprocity,
> subject-benefactive, subject as agent+experiencer.

It may be that I have had in mind in the case of BAPTIZW/BAPTIZOMAI  
is the type with patient-beneficiary functions for the object but I'm  
thinking of what is commonly referred to in the grammars as the  
"causative middle":

Smyth ¤1725. "The Causative Middle denotes that the subject has  
something done by another for himself: EGW GAR SE TAUTA EDIDAXAMHN  
'for I had you tught this' Xen. Cyroped. 1.6.2, PARATIQESQAI SITON  
'to have food served up 8.6.12, hOSOI hOPLA AFHiRHNTAI, TACU ALL  
POIHSONTAI 'all who have had their arms taken from them will soon get  
others made 6.1.12, hEAUTWi SKHNHN KATESKEUASATO 2.1.30. [This last  
does indicate beneficiary with a reflexive pronoun, of course. Smyth  
also notes that this force does not belong exclusively to the middle  
and points to a similar discussion of a Causative Active in ¤1711.]

BDF ¤317 (p. 166) The middle in the sense of 'to let oneself  
be ...' (cf. German _sich lassen_) (causative; cf. °314) also occurs  
in scattered passages in the NT:
'get themselves emasculated' (cf. APOKOMMENOS LXX Dt. 23:1;  
PERITEMNESQAI on the other hand is treated as a pass.). EBAPTISAMHN:  
Acts 22:16 BAPTISAI KAI APOLOUSAI (1 Cor. 6:11 APELOUSASQE) can be  
explained as causal; but in 1 Cor. 10:2 -ISANTO appears to be  
spurious (KLP; in p46 corrected from EBAPTIZONTO) and -ISQHSAN alone  
to be correct; EBAPTISQH in Lk 11:38 in an entirely different sense  
'to dip his hands' is incorrect (EBAPTISATO is correct with p45  
minusc. 700). APOGRAFESQAI Lk 2:1, 3 also belongs here 'have oneself  
enrolled' on account of the aor. -GRAYASQAI in v. 5. -- Mayser II 1,  
89; 109, Robertson 808f.

BDF ¤314 (p. 165) The passive in the sense of 'to allow oneself to  
be ...' (cf. German _sich lassen) was common from earliest times:  
ADIKEISQE 1 Cor 6:7 'let yourselves be wronged' (in the sense of  
allowing it), likewise APOSTEREISQE in the same vs. BAPTIZESQAI 'get  
oneself baptized' (in the sense of to cause it; aorist EBAPTISQHN,  
cf. however ¤317.
	Permissive: DOGMATIZESQAI 'submit to regulations (lit. let yourself  
be regulated)' Col. 2:20, GAMIZESQAI (¤101), SUSCHMATIZESQAI and  
METAMORFOUSQAI Rom 12:2, hILASQHTI 'let yourself be disposed to  
grace' (lk 18:13. Causative: hAGNISQHNAI Acts 21:24, 26 DIAKONHHNAI  
MT 20:28=MK 10:45, often PERITEMNESQAI (also in the sense of allow).  
Cf. Robertson 808f.

Robertson does indeed deal with this in his big grammar, 808f., but I  
won't cite that.
Guy Cooper (1998 revision of Kruger), Attic Greek Prose Syntax:  
Revised and Expanded in English, Volume 1 (Attic Greek Prose Syntax),  
¤52.11.0: "The middle is used to show an action which is carried out  
at the command or behest of the subject"; ¤52.11.1 "This force is  
simply a matter of conception É understood without difficulty in ages  
of widespread and often easy subordination. It is a causal use of the  
verb which is not essentially different from comparable uses of the  

I'll have more later in a part 2 on the lexical matters and usage of  
BAPTIZW/BAPTIZOMAI in response to Iver's Oct. 20 message.

Carl W. Conrad
Department of Classics, Washington University (Ret)

More information about the B-Greek mailing list