CJ-ONLINE Archives

April 2009


Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Reply To:
Classical Journal On-Line <[log in to unmask]>
Sun, 5 Apr 2009 10:09:35 -0500
text/plain (198 lines)
The Moods of Homeric Greek. By JO WILLMOTT. Cambridge Classical Studies. 
Cambridge and New York: Cambridge University Press, 2007. Pp. xii + 264. 
Cloth, $99.00. ISBN 978–0–521–87988–0.

Order this text for $91.49 from Amazon.com 
using this link and benefit CAMWS and the Classical Journal:


Previously published CJ Online reviews are at 

-Transliterated Greek has been set off within double asterisks
-Aspiration and accent (in that order) follow their vowel

CJ Online 2009.04.01

Latin has a sequence of tenses, Greek has a sequence of moods: so goes the 
standard line. And Classical Attic is relatively well-behaved in using 
subjunctives after non-past main verbs, optatives after past-tense 
ones—with the proviso that an author could always choose in the latter 
instance to use the “vivid” subjunctive, whatever that is supposed to 
mean. This distribution, together with the use of main-clause subjunctives 
in exhortations and prohibitions (over which the speaker has some control) 
and that of optatives in wishes (over which the speaker has little 
control), led scholars to posit that the two moods act in parallel, and 
that they exist, together with the indicative, on an irrealis continuum: 
with the indicative, the speaker asserts the reality of the event, with the 
subjunctive that assertion becomes an expression of will, with the optative 
one of mere wish. There were also morphological grounds for this schema: 
the subjunctive has primary endings, the optative secondary endings, 
calling to mind the pairing in English of will : would, can : could, and 
may : might, in which the past tense of the modal verb is more irrealis 
than the present.

Nor is this irrealis continuum the only major theoretical categorization of 
relevance to the Greek moods. Perpendicular to it runs a division of moods 
and modal verbs into deontic and epistemic modality. As the names suggest, 
the former covers the use of the moods to indicate obligation, necessity 
and will (prohibitive and jussive subjunctives, the optative of wish), 
while the latter encompasses modal uses in which the focus is more on the 
speaker’s uncertain knowledge of the truth of the statement (potential 
optatives, and, in Homer, the subjunctive used as a future). Again, 
objective morphosyntactic facts seem to corroborate this division: as 
Chantraine pointed out, deontic modals generally lack the modal particle 
**a)/n** / **ke** and are negated with **mê/**, whereas the epistemic 
modals have the modal particle and are negated with **ou)**. As a whole, 
the description appears to hold fairly well for Attic. But what about 
Homeric Greek? The central argument of W.’s book is that it does not. 
Instead, W. offers a problematized account of the Homeric moods, in which 
these neat structuralist divisions are called into question and replaced 
with a more complicated network of usages she sees as ultimately derived 
from earlier Ur-meanings through grammaticalization. (For the uninitiated, 
grammaticalization is the process whereby words of semantic weight, like 
English will, gradually become bleached of semantic content and, frequently 
shortened to clitics—’ll—and affixes, come to serve as markers of 
grammatical features like tense or mood.) Because of the tenacity with 
which W. argues against received opinion, the book will no doubt become 
essential reading for those interested in the Greek moods. But it should 
not be taken as the last word on the subject: not all the problems W. sees 
in the standard view are of equal weight, and, more importantly, as W. 
would herself agree, this is simply too vast a topic to be given definitive 
treatment in 200-odd pages.

After two chapters covering the theoretical background in good detail (pp. 
1–36), W. begins the body of the work with a chapter on the indicative 
(pp. 37–52), setting the base-line for the examination of the subjunctive 
and optative, which are her central focus. W. argues against the position 
that the indicative is either (a) particularly realis or (b) epistemically 
neutral, suggesting instead that it signals a “positive epistemic 
stance.” In contesting position (a), she follows F.R. Palmer’s line 
that the ability of indicatives to collocate with adverbs like possibly 
shows that they are not realis forms; but neither does she agree with 
Palmer that they are epistemically neutral, pointing to the incompatibility 
of the indicative with modifiers like doubtfully. In W.’s view, the 
reformulation of the indicative as a marker of positive epistemic stance 
also explains the most modally troubling uses of the indicative, in 
counterfactuals. But difficulties remain. First, it is unclear exactly how 
“positive epistemic stance” and “realis” differ as descriptions of 
the indicative. W. has already pointed out (p. 14) that the term realis 
refers not so much to objective reality as to the speaker’s presentation 
of the proposition. Clear examples of clauses that would be realis but do 
not have positive epistemic stance, and vice versa, would have been welcome 
to elucidate the distinction W. is attempting to make. Presumably we are 
meant to believe that Tom is possibly singing would be an indicative that 
shows positive epistemic stance but not realis modality. But if the 
presence of possibly is enough to prove false the interpretation of the 
indicative as realis, is it not also enough to refute the positive 
epistemic stance reading as well? Second, the counterfactuals are still 
awkward: just as with a description of the indicative as realis, it remains 
problematic to argue that a counterfactual indicative in an apodosis 
represents a more positive epistemic stance than does the potential 
optative of the future less vivid.

Chapter 4, on the subjunctive, is the longest in the book (pp. 53–112) 
and is divided into three main sections dealing with the chief uses of the 
mood: the epistemic (Monro’s quasi-future subjunctive), the hortative and 
the negative directive. In the first, W. examines the difference between 
the future indicative and subjunctive as markers of future events. There is 
much overlap between the two, as is only to be expected given the similar 
situation with future markers in modern languages, e.g. English will and 
going to. On the basis of such pairings as **ou)d’ e)/ssetai ou)de\ 
ge/nêtai** (Od. 16.437), W. does not believe that the future signals more 
certainty about the upcoming event than the subjunctive does. Instead, as 
the chief difference between the two, she points to the strong preference 
for the subjunctive over the future in conditional and temporal clauses 
referring to the future. W. convincingly ascribes this distribution to the 
fact that the future is a younger marker than the subjunctive. Assuming, as 
seems safe, that the Indo-European subjunctive was a future marker before 
the sigmatic futures seen in Greek arose, we can see in the Homeric 
situation the expected distribution of the older marker in more modal 
contexts, the younger one restricted (in subordinate clauses) to resumptive 
conditionals and indirect questions. This pattern has a parallel in 
Spanish, where the older synthetic future has more modal uses than the 
younger periphrastic forms. As for W.’s study of the negative directives, 
her chief point is the refinement of what might be called the 
Meid–Hoffmann model, which distinguishes between one form (in Greek, the 
present imperative) as prohibitive—don’t do this thing you’re already 
doing!—and another (here, the aorist subjunctive) as preventive—don’t 
start doing this thing you’re not doing yet! [[1]] To this, W. adds the 
idea of control: with present-imperative prohibitives, it lies in the power 
of the addressee to stop the action in question, whereas such action lies 
outside the addressee’s control in the aorist-subjunctive preventives. 
Readers may not agree with all W.’s examples—the three passages with 
**nemesa/ô** on pp. 102–3 suggest, as she apparently admits, that the 
chief determining factor is temporal—but control remains a useful 
parameter to keep in mind when examining negative directives.

This is a concept W. returns to in her next chapter, on the optative (pp. 
113–52), where she concludes, inter alia, that the difference between 
optatives of wish and imperatives is not that the former are weaker than 
the latter, but that they lack the element of control on the speaker’s 
part. More problematic is the material in the first half of the chapter, 
where W., eager to confound the notion that the optative is a less realis 
version of the subjunctive, suggests that it should instead be portrayed as 
indicating a negative epistemic stance and thus forms a closer pair with 
the indicative. This position is supported largely by W.’s analysis of 
conditionals, in which she argues that the optative in protases and 
apodoses is not always a remote possibility, but rather presents events as 
unreal. Apart from the fact that it is unclear how this description is to 
be distinguished from the standard position that the optative is the most 
irrealis mood (is W. really correct in saying on p. 122 that it is “just 
coincidental that ‘unreal’ events will often be ones with a smaller 
likelihood of happening than real ones”?), W.’s interpretations of the 
conditionals are themselves subject to question: she holds that the 
optative does not in fact occur in particularly remote conditionals, yet in 
her examples on pp. 116–23, she nearly invariably translates it with the 
past-tense protasis and would apodosis that is, to my mind, precisely how 
English indicates remote possibility (contrast the present-tense protasis 
and will apodosis of “more vivid” conditionals). True, some of these 
optative protases may be relatively likely to be fulfilled, but there might 
be pragmatic reasons for presenting them as only remotely likely. [[2]]

Generally more satisfactory is Chapter 6, which treats the use of the moods 
in purpose, iterative temporal and non-specific relative clauses (pp. 
153–91). W.’s position that Homeric Greek has not yet reached as rigid 
a sequence of moods as Classical Attic is a reasonable one, and, 
considering the numerous exceptions that arise if one pretends that it has, 
W. must be right to see the choice of the moods as determined through the 
semantics of the subjunctive and optative independent of their relationship 
to the main verb. The book concludes with a brief summary (pp. 192–8); a 
justification of the decision not to treat the modal particle or choice of 
negative as important (pp. 199–210; some may still want to defend 
Chantraine’s position); a complete list of the lines of Homer that 
exhibit the constructions she discusses (pp. 211–37); a very full 
bibliography, especially rich in theoretical linguistic literature; and 
indices of passages and topics covered. Overall, while readers may disagree 
with W. on some points, they will no doubt be stimulated into reconsidering 
exactly what the Homeric moods do: Attic this certainly is not.

University of Virginia
[log in to unmask]

[[1]] Those interested in the corresponding problems with negative 
directives in Latin can now turn with profit to Chapter 4 of W.D.C. De 
Melo, The Early Latin Verbal System: Archaic Forms in Plautus, Terence, and 
Beyond (Oxford, 2007).

[[2]] For example, the optatives in Il. 5.273 that W. says refer to a 
relatively likely event (p. 119) could be deliberate understatement (It’s 
likely, but I’ll pretend it isn’t), perhaps out of a desire on 
Diomedes’ part not to appear too cocky in going after horses that were 
descended from those given by Zeus to Tros in exchange for Ganymede. Even 
more to the point, the capturing of the horses is itself contingent 
on—and thus more remote than—the successful killing of Pandarus and 
Aeneas, an act marked as dependent on the fulfillment of a future more 
vivid protasis (**ai)/ ken** … **o)re/xêi**) in lines 260–1. Here, at 
least, the Homeric moods follow the Attic playbook.

You may remove yourself from the CJ-Online list-serv 
by sending an email to: [log in to unmask] 
Leave the subject line blank, 
and in the first line of the message write: UNSUBSCRIBE