Sunday, August 9, 2015

Romans in Reconstructed Koine

Nine years ago, as I prepared to teach Paul's letter to the Romans for the first time, I recorded the Greek text so that I could listen to it on the weekly commute I was doing that year to and from Saskatoon. I found I could make it through the entire letter on the two and a half hour drive, and the experience of listening to the whole thing more-or-less each week did a lot to give me a sense for thematic connections within the letter, as well as the flow of thought in the letter as a whole. But the quality of the recording (and of my newly-adopted Reconstructed Koine pronunciation) was so bad that I could not in good conscience share it with anyone else.

I recently finished recording Romans again, and this time I am not as embarrassed by the result. As far as I know, it is the only recording of the Greek text of Romans that uses the "Reconstructed" or "Imperial" Koine pronunciation system that most closely approximates how Greek was spoken in the first century.

  • My pronunciation follows the Reconstructed Koine system as formulated by Randall Buth in his "Notes on the Pronunciation System of Koiné Greek" (online here; my simplified summary is here). The one (intentional) exception is that I typically preserve the rough breathing, a decision that Buth allows for in a footnote:
"Students are free to add aspiration as they wish, though one may imagine that such would have been thought stuffy or snobbish in the first century. There may still have been some features of a classical Greek that were consciously learned by the upper classes and in which [h] would be learned and heard." (Buth, footnote 24)
My current practice, reflected in these recordings, is to omit the rough breathing when it sounds particularly odd, is difficult to pronounce, or makes little difference in sound (e.g., οἷς, ὁμοθυμαδόν); but to retain it in other instances because the 'h' sound frequently distinguishes between two different forms that would otherwise sound alike.
  • I concentrated on getting the pronunciation and accents right, and, less successfully, on meaningful phrasing. Focusing on these elements meant that I was unable to give as much attention as I would like to reading with inflection. I am also afraid there are still unnatural pauses before words about whose pronunciation I was not confident, particularly in the first few chapters. So the result is still far from a professional recording, but I hope it is serviceable. (If you are simply interested to hear how Reconstructed Koine should sound, I recommend the samples on the Biblical Language Center's website here.)
  • The quality of recording improved as I went along--so much so that I re-recorded chapter 1 at the very end; chapters 2 and 3 should probably be redone as well, but I didn't want to get caught in an endless loop on a project that had already consumed more time than I had to give it. 
  • I would be happy for any suggestions, feedback or technical advice, in the event that I try to do something like this again.
You can listen to individual chapters below or download all the files here (wav) or here (mp3).

Romans 1

Romans 2

Romans 3

Romans 4

Romans 5

Romans 6

Romans 7

Romans 8

Romans 9

Romans 10

Romans 11

Romans 12

Romans 13

Romans 14

Romans 15

Romans 16

Creative Commons Licence
These Romans in Reconstructed Koine recordings by David M. Miller are licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. This means you are welcome to download, copy and use the recordings with attribution, but you may not modify them without my permission, or sell them.


Unknown said...

A huge improvement on the Erasmian pronunciations. Personally, I don’t think the Buthian pronunciation goes far enough. With respect to the First Century (and before), the evidence suggests that the current pronunciation of Greek remains our best guide. Most recently, this has been demonstrated by Caragounis (, Zachariou (, and Theophilos ( Not only is it true to the period, but it is more advantageous for textual criticism and facilitates an immediate and direct transfer to Modern Greek. I wonder why you did not opt for this pronunciation?

d. miller said...

Thanks for your comment, Jody. One of the advantages of Reconstructed Koine is that it makes the transfer to modern a lot simpler, and there are a lot more free audio resources available using the modern Greek pronunciation. For example, I can now listen to and appreciate the excellent recordings of the LXX and GNT in modern Greek, as well as the liturgy of St. John Chrysostom in Byzantine chant.

There are two main reasons why I chose not to adopt modern Greek pronunciation: The first is that, despite Caragounis, etc., I do not take it as demonstrated that η and υ had itacized to ι before the 2nd century CE. For evidence, I recommend Buth's discussion in the article I mention in my post (see esp. p. 221f.). Buth's discussion is based on Allen and especially Gignac, so it is not exactly fair to call it 'Buthian'; he also interacts with Caragounis. See also Daniel Streett's summary of a 2011 SBL discussion:

The second reason is pedagogical: English-speaking students tend to find the German ü difficult to pronounce, but a 7-vowel system results in fewer homonyms, which simplifies the language-learning process. (This is also why I retain the rough breathing.)

Unknown said...

Many thanks for that, it is always interesting (to me at least) to learn about people’s reasons (or more often, lack thereof) for following one pronunciation over another. I agree that Buth’s reconstructed first-century Koine pronunciation – I will continue to refer to it as “Buth’s” for clarity, and because Allen is attempting to reconstruct a Fifth Century BCE literary Attic pronunciation – is more advantageous than Erasmian with respect to transferring to Modern Greek, but the “modern” pronunciation is of course even more so.

I have just reread Buth on η and υ. With respect to υ he demonstrates that it is interchangeable with οι but he doesn’t demonstrate that the sound was (IPA) y; this is taken for granted, and overlooks the evidence for the interchangeability of υ and ι, and οι and ι in the early Hellenistic period (set out in Caragounis, Development of Greek, 367-70). In other words, it looks like υ and οι are interchangeable because they were both pronounced like ι.

Similarly, there is significant evidence for the interchangeability of η with ι/ει in the early Hellenistic period (see Caragounis, 370ff.). Buth accepts the method and evidence provided by Caragounis with respect to the Second and Third Centuries CE but ignores what he says about the earlier Hellenistic period, even though the method employed and nature of the evidence is the same.

We can never know for sure of course how exactly the ancients pronounced Greek, and because of this I do respect Buth’s reconstruction. I accept that it may well represent the way some Greek speakers sounded around the turn of the Common Era, but, at the very least, the evidence suggests that it is not representative; there were those for whom ι=ει=η=υ=οι, and that if this was not widespread by the turn of the Common Era, it very soon was, and remained so to the present day. Anyway, I would be very interested to know why you do not trust the evidence and argumentation provided by Caragounis et al.

d. miller said...

Hi Jody,

Thanks for your follow-up comment. Two quick notes:

(1) Buth does provide an example (admittedly only one) for the equation of υ and IPA /y/. See p. 221 pair 4 b.

(2) For an extensive discussion and critique of Caragounis's view that η had already itacized with ι/ει, see this b-greek thread: I have nothing to add to that discussion.