The Middle Voice Panel Discussion: An Interim Comment

Buist Fanning’s review of The Linguist as Pedagogue: Trends in the Teaching and Linguistic Analysis of the Greek New Testament edited by Porter & O’Donnell for the RBL has been made available at the RBL Website.

Of note for our purposes are Dr. Fanning’s comments about Pennington’s contribution to that particular book:

Jonathan T. Pennington’s “Setting Aside ‘Deponency’: Rediscovering the Greek Middle Voice in New Testament Studies” is a must-read for Greek teachers. He presents a strong and clear argument that the traditional description of Greek middles and passives as “deponent” is erroneous and should be abandoned, because it reflects a misunderstanding of voice usage, especially of the middle. He marshals some important evidence and clarifies several significant points in his discussion, but his conclusion is ultimately too extreme. He argues actually against a simplistic view of deponency and fails to come to terms with a more nuanced approach such as that found in the recent collection of essays edited by Matthew Baerman and others (Oxford University Press, 2007). Deponency is a valid category in New Testament Greek, but Pennington has helped clear away some common misunderstandings of it.

It should go without saying if you’ve been following this site for any amount of time that I thoroughly disagree with Dr. Fanning that Pennington’s view is too extreme. I also disagree with his statements about Baerman et al. Deponency and Morphological Mismatches (Proceedings of the British Academy). The Baerman volume isn’t a nuanced approach because it is dealing with a very different question: What are the implications for linguistic theory if mismatches between form and meaning exist? The question of whether Ancient Greek has such mismatches in its diathesis system is very much open to debate. I have already argued else where that Ancient Greek speakers did not believe that to be the case (“Dionysius Thrax & Translating Πάθος“).

The only portion of the Baerman volume that touches on the question examines Greek diachronically (Lavidas & Papangeli’s “Deponency and the Diachrony of Greek”) and refuses to engage the participants of the discussion. Rutgar Allan and M. H. Klaiman are not mentioned at all, while Suzanne Kemmer, Linda Manney and others are dismissed with a paragraph each–not to mention Neva Miller, Bernard Taylor, Carl Conrad, and others who wouldn’t be on their radar. As a Classicist, perhaps the unwareness of Allan could be excused, but to not even mention Klaiman (whose monograph was published one of the most respect linguistics monographs series in existence) cannot be justified. It can be hardly said that such a volume presents a “nuanced” view. The most nuanced discussion in the volume is that if P. H. Matthews entitled, “How Safe Are Our Analyses,” where he argues that the morphological mismatch in Latin that produced the term deponency isn’t as neat and tidy as everyone wants it to be.

With that said, I completely agree with his statement that Pennington’s article presents too simplistic a picture. This is a natural result of Pennington’s work arising from his own reading of the text rather than an thorough analysis. Fanning”s criticism here would not hold weight against the more detail (and monograph-length) work of other scholars.

I have the highest respect for Dr. Fanning, but I reserve the right to disagree with him.

20 thoughts on “The Middle Voice Panel Discussion: An Interim Comment

  1. I see that my email to you this morning was stale before I wrote it — at least as regards this review. It’s sad that some horizons are as narrow as they are.

  2. I really wanted to get this book, but its hard justifying spending that much money on it. The chapter on relevance theory makes it enticing though.

    1. Its a good book, though (as always) there are some terrible articles and some very good ones. The diachronic study of Greek in it is pretty bad. Baerman’s introductory article is excellent, though according to Carl (I e-mailed it to him), the perspective is thoroughly grounded in traditional Latin grammar and also it fails to even dialogue with those who have put forward alternative proposals (e.g. Kemmer).

      1. Here’s my very brief evaluation of the article on “Deponency in the Diachrony of Greek” by Lavidas and Papaneli, for what it’s worth:

        The authors have undertaken a task that is too large and too important to deal with in the rather cursory fashion that they have employed. Several tables of usage of “deponent verbs” over the course of Greek linguistic history are offered, but it’s troubling

          (a) that there is no statistical validation of the morphological voice-forms in each list,
          (b) that the Greek verbs are listed in an inconsistent transliterated Roman-letter scheme rather than in a Unicode Greek font, and
          (c) that the one-word English semantic equivalent offered for each form is too often questionable or misleading, but most of all,
          (d) there is all too little effort made to make sense of the data presented in these tables.

        When it comes to the task of explaining the observed fact that there is a great deal of persistence of middle-passive morphology for so many of the verbs whose history has been followed over the course of Greek linguistic history, the authors offer a very brief review of three possible explanations; they quickly dispense with the semantic explanation, that these verbs fall into readily-discernible categories of subject-affectedness, with a brief reference to the work of Kemmer and Zombolou (whom I haven’t read). They also reject, more reasonably, a syntactic explanation, that linkage to certain kinds of complements might account for middle-passive morphology. They turn ultimately to a morphological explanation: “We take here the morphological approach to allow for a non-systematic view of the data, in the sense that the feature specification that accounts for deponent verbs is realized in a random way and possibly relies on the idiosyncratic properties of each verb.”, p. 120 — I do think that there’s a little bit of truth here: voice forms and usage are idiosyncratic to those everyday verbs that have retained more archaic morphology and not been brought into conformity with standard patterns of morphology; that is why ἔστην and ἕστηκα are the aorist and perfect forms of ἵσταμαι. But the failure of the authors to take seriously the categories described by Kemmer and at least hinted at by Zombolou (whose work I haven’t seen) marks this paper that might have had some value as a failed effort to make sense of some useful data that has been assembled but not interpreted in any really meaningful way.

        1. Carl, I totally disagree with your remarks. I think that an article could not contribute to the discussion of verbs and voices if it just follows Kemmer’s categories (that actually seem not to offer a clear explanation of the data presented in the paper you discuss). The positive aspect of the article you mentioned is that it gives us a different (new) analysis of some data that Kemmer didn’t deal with.

        2. I think that an article could not contribute to the discussion of verbs and voices if it just follows Kemmer’s categories (that actually seem not to offer a clear explanation of the data presented in the paper you discuss).

          Then you simply don’t know the literature. It’s one thing for Lavidas and Papaneli to have a different view than Kemmer (because *everyone* does–nobody has done what you’ve said here). It’s a completely other thing to brush off those with whom you disagree in a paragraph and not actually deal with their claims. If you think that’s a good thing, then you have a strange idea about what scholarship should look like.

          Further, Lavidas and Papaneli fail to engage in the most important piece of secondary literature on this topic for their purposes: Rutger Allan’s The Middle Voice in Ancient Greek:.

          Lastly, different//new analysis are only a positive if they stand up to rigorous scrutiny with regard to the evidence. Lavidas and Papaneli fail in that respect on a number of points.

      2. The diachronic study on Greek deponents is worth reading since it is one of the very few linguistic papers that discuss diachronic (not pseudo-diachronic, that is, just different synchronical periods) aspects of deponency.

        1. Yeah…to bad they botched to so badly.

          They give no *evidence* for the *meanings* of the verbs in question or the valency of the verbs in question. It’s all just a pile of lists. They claim that they only deal with transitive deponents, but use numerous verbs that are demonstratably *not* transitive. They fail to deal well with dialectal evidence and often treat an active from from one dialect as if its the counterpart to a middle-passive from another dialect.

          I think your idea of diachronic research is quite odd. If anything, I would say that you have it turned around. Diachronic research that either does not or cannot give an account of a language as a synchronic system at a given point in the language’s history is the true pseudo-diachronic.

  3. Mike, thanks so much for bringing the SBL discussion of the middle verb to those of us who could not attend. I appreciate your insightful interactions too. I’ll be involved in teaching three different Greek courses this year in PNG, and these questions regarding the middle verb always come up. Two years ago, I had Carl Conrad’s thoughts on the matter, and it looks like a consensus is developing in that same direction. That’s good to hear. Are the SBL papers available somehow or do I need to contact whoever presided? Who was it, Stanley Porter? Not sure how free they are for distribution after the meeting.

    1. Randall Tan was presiding over the panel. The panel consisted of Porter, Pennington, Taylor, & Campbell.

      I think that the plan is to get the papers published…which doesn’t help you a whole lot. It might be worth contacting the authors, they might be willing to share with you.

        1. Oh my god Mike!
          I think you couldn’t read the paper well.
          You say: “They fail to deal well with dialectal evidence and often treat an active from from one dialect as if its the counterpart to a middle-passive from another dialect.”
          Please read the specific lines in the paper again!
          Have you written something on the subject?
          Or you just talk on internet???

        2. Peter,
          You’re welcome to comment as long as you stop insulting people. Is that really necessary?

          Here’ a good way to dialog with people. You might have said something like this:
          “Mike, are you sure you got that right? Maybe you could cite some evidence as to what you’re talking about. I’m not seeing the same thing.”

          And then I might have replied:
          “My goodness! You’re almost right, I dd made a mistake. What I intended was not a dialectal difference, but a genre difference. I confess that I was speaking from my memory of the article and did not have the book in front of me at the time. Big mistake on my part!! I’m home now and have pulled the book off the shelf. So let me correct my error: They write as if θιγγάνω and ἅπτονται are identical in meaning in order to justify their claim that deponency cannot be linked to a particular semantic reading (page 103). But their examples are from prose (ἅπτονται) in one and poetry (θιγγάνω) in another. But going back and looking through the article, I have actually found another example with problematic use of dialects on the very next page. I’ll quote the paragraph at length and then comment on it:

          “Another example is provided from the comparative perspective of dialectology. From a cross-dialectal perspective, the active verb katalambano:, from the Attic Classical period, and deponent verb katalambanomai, from the Ionic Classical language, are used in the same contexts with the reading ‘I occupy’. Once again, the deponent verb and the active verb have identical interpretation, which necessarily rules out any particular semantic function of the middle/passive morphology of deponents” (104).

          I honestly do not know what to do with such a claim. Because they are separate dialects, we shouldexpect their speakers to have different mental lexicons. Exactly why should this be evidence for their claim? They try to use dialectal evidence without accounting for dialectal difference.

          I’m currently waiting for Lavidas’ dissertation to arrive in the mail. I’m hoping that he’ll have more argumentation there than in this survey/list of lexical items we have here.

          Other problems with their article include their listing of “apekri-same:n” as the middle past of ἀποκρίνομαι, when the actual middle past form of this verb is ἀπεκρινάμην (page 100). That’s just sloppy and they have exactly the same problem for the middle future of this verb.

          In the section of Hellenistic/Roman texts, they make the same claim about no difference in meaning between actives and middle/passives, but they want the claim to stand on a single example from Mayser. Seriously?

          Then they say, “The distinction between middle and passive suffix in past and future tense, that had started to disappear already from Ancient Greek, disappears completely in the Hellenistic Koine.”

          This is only “true” if you assume that the meaning of -θη is passive. But Rutger Allan has already demonstrated that it isn’t going to Homer. Even in Homer, -θη was already being used for mental process middles, body motion middles, collective motion middles, and spontaneous process middles well beyond the supposedly canonical passive (Allan 126-147). All that can really be said about the -θη suffix is that it at one point in its history expressed only a subset of middle usages, but over time took on more and more in the aorist and future. By the Classical period, it was encroaching on perception middles, mental activity middles, speech act middles, and reciprocal middles (Allan 148-156). In the Hellenistic and Roman period, it simple became the central marker for middle meaning in the aorist and future.

          Anyway, they continue:
          “Neither this change nor the loss of hte middle reading (involvement of the subject) had implications for deponenct verbs. Conversely, a number of ‘novel’ deponent verbs are attested in the Hellenistic Koine period of Greek (Mayser 1970), as illustrated in the Table 7.”

          Here are their verbs (I can connect all of them to a semantic middle meaning) and some of them aren’t novel at all (supposedly, these verbs appear for the first time in the Hellenistic Koine).
          enaraomai ‘I adure by’ = speech act middle (like μυθέομαι or ἐνεύχομαι)
          ekkarpizomai ‘I reap, I enjoy’ = indirect reflexive middle
          drassomai ‘I grasp with the hand’ = indirect reflexive middle already used in Herodotus 3.13 ταύτας δρασσόμενος αὐτοχειρίῃ διέσπειρε τῇ στρατιῇ.
          sphragizomai ‘I close or I enclose with a seal’ = indirect reflexive–i.e. it shouldn’t be glossed, ‘I enclose with a seal’, rather it should be glossed as ‘I enclose with my seal.’
          deksioomai ‘I canvass, I greet with the right hand’ (used as early as Xenophon) = body motion middle.
          epilanthanomai ‘I let a thing exacpe, I forget’ (used in the Classical period) = spontaneous process or mental process middle (depending on the sense).
          eksidiazomai ‘I win over’ = indirect reflexive middle. This is an interesting one. Their gloss is clearly from LSJ, but they conveniently choose the gloss that fits their purposes. The other two suggested glosses are: ‘appropriate for oneself’ and ‘receive for one’s own use’–not semantically determined, eh? Right…
          ananeomai ‘I mount up’ = body motion middle. This is in the Odyssey 10.192! Hellenistic indeed! οὐδ’ ὅπῃ ἀννεῖται ‘nor where he rises;’
          diomologeomai ‘I agree mutually’ There are numerous example from the Classical period of this middle in LSJ–which, incidentally, is where they get their gloss. The mutually should have been the clue. This is a reciprocal middle.
          theoroemai ‘I look at, I view’ = Perception middle.

          There nothing in these verbs that demonstrate that middle verbs are not semantically determined. If they have even skimmed Allan, they’d know this. All of these can be subsumed in the abstract conceptual schema of subject affectedness, which as I’ve already noted in a previous post in this serious is exactly how we should probably understand πάθος in Dionysius Thrax. Allan’s Radial Network for the Middle voice in Ancient Greek demonstrates the semantic underpinnings of the Middle from Homer through the Classical period and there’s nothing in the Hellenistic period that contradicts that. For a brief summary of Allan approach, you can read this post that I’ve written here: Clarifying Allan & Kemmer on Middle Voice: Cognitive Linguistics.”

          Anyway, so that’s how I would have replied had you not insulted me.

        3. Dear Mike,
          It is more insulting to write things that do not exist or you don’t remember well, as you admit in your reply to me.

          As I see, the main problem here is the problem you may have with Generative Grammar.

          If you don’t like a framework, just say it clearly, but don’t write things that do not exist (or in the way you’d like to interpret or remember).
          Peter

  4. Peter,

    I apologize for insulting you. I admit that I was curt and rude in my response to you. I may have been offended by your words, but that gave me no right to respond the way I did. If you’d accept my apology, I would absolutely love to discuss the data from the article with you.

    If I may, I would ask why you think generative grammar is the issue here? I must confess I am at a total loss on that point since Lavidas & Papangeli say nothing about the linguistic framework their working in. Everything I said above relates to the linguistic data. Besides, I am a generative linguist–in spite of the fact that I occasionally dabble in Cognitive linguistics, but that’s a natural result of the fact that my wife is a Cognitive linguist.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s