The Stoics’ notion of a lekton is a well-known peculiarity of their logic and philosophy of language. Why they thought it was necessary to introduce what is commonly translated as a ‘sayable’ is less easily established and has been a source of controversy and confusion since antiquity. In this book Alessandrelli tackles the question straight on without ultimately dispelling either the controversies or the confusion — though there is an attempt at cutting some sort of middle path between the two most important of the conflicting interpretations.
The book has four chapters. In Chapter One, the author gives a developmental account of the Stoic lekta: Cleanthes called ‘sayable’ those items which Zeno spoke of as predicates and propositions on the grounds that they are semantic items. The author argues that lekta are semantically dependent on causation (they describe the attributes acquired by bodies) rather than constitutive of it — in stark opposition to a view according to which lekta are real items in the world, the real effects of causes (a view put forward by Michael Frede). Chapter Two focuses on the notion of a semantic correlate by highlighting the ties between language and thought and analysing language as the result of a mental process in which rational impressions are articulated by means of lekta. Chapter Three revisits the notion of complete and incomplete lekta, repeating some elements from Chapter One with added support from the discussion from Chapter Two. The commitment to the view that lekta are dependent correlates of words is emphasised and leads to claims that predicates cannot be the only kinds of incomplete lekta. The last chapter is devoted to the Stoic treatment of cases, ptôseis, again echoing statements from Chapter One, claiming that cases are signifying expressions — and not signified like lekta — which cover inflected forms of all words which are not, grammatically speaking, verbs.
A crucial question concerning lekta is whether they are fundamental to Stoic philosophy or whether they are an anomaly, a loose end which it is the interpreter’s task to somehow tidy up. The very title of this book is indicative of the reading favoured by the author: lekta are a problem and some acrobatics are therefore required in order to save the Stoics from seemingly incomprehensible clumsiness. The author’s main claims are that lekta are incorporeal meanings of words; that these lekta by their very nature need to enter into combinations, require saturation as it were; and that such combinations are made possible by the ordering of words into significant speech. Sayables are thus derivative entities dependent on language. Alessandrelli explains their presence on the grounds of an argument, which he claims without further demonstration was taken by the Stoics themselves from Plato’s Euthyphro. He suggests that the Stoics transposed Socrates’ proposition that ‘what is seen is seen because first of all I see it’ (10b-c) into: ‘it is because we are saying something, that something has been said.’ Thus a lekton is fundamentally a semantic item utterly derivative on words and sentences. Alessandrelli follows a certain line of interpretation whose antecedents in the second half of the 20th century are in Graeser, Rist and to some extent Long – although the unmotivated appeal to the Euthyphro is, to my knowledge, a novel twist on the more analytical arguments put forward by A. Graeser (‘The Stoic Theory of Meaning’ in J. M. Rist, ed., The Stoics, Berkeley 1978, 77-100).
A first reaction to Alessandrelli’s view is that it renders lekta semantically redundant: what do they contribute? They are not external references of words or sentences since they have no independent status and are not corporeal; they rather appear as utterly dependent on surface grammar. What is more — a point which Alessandrelli ignores — this reading would need to provide a substantial explanation of the meaning (which is what lektaare considered to be) of prepositions and conjunctions; an enterprise which the earlier critiques of the Stoics were indeed forced to delve into (e.g. Ammonius, Commentary on Aristotle’s De interpretatione 12.27 ff., where Stoic criticism on the redundancy of lekta is merged with Peripatetic concerns about the parts of speech — conjunctions and articles here in particular).
This reading furthermore greatly diminishes the one rather clear element of doctrine we have in our texts, namely the clear-cut distinction between corporeal utterances and what these expressions manage to get said, namely lekta. The distinction is suggestive of some form of independence of the one from the other, or at least requires some consideration rather than an assertive but over-hasty account knitting word and lekta inextricably together and thereby blurring from the start the grounds for the distinction. Lekta are repeatedly listed by the Stoics as one of only four incorporeals, together with time, space and void, and yet there is not a line in the book attempting to give a satisfactory account for this tetrad, which has such strong textual evidence.
The bulk of Alessandrelli’s account lies with his analysis of the relation between rational impressions and lekta. He focuses on one of the main descriptions of lekta from the doxographies as the kind of items which “subsist according to a rational impression”. He comes up with a description of a lekton as that which enables the content of our impressions to be articulated into meaningful speech. The author’s main worry here is not to assimilate the very content of an impression with a lekton. It is the content, which is not an incorporeal lekton, which has a corresponding lekton whose purpose is to enable us to say what that content is. There is some consistency in this suggestion with the first semantic dependence claim mentioned above: lekta are consistently relegated to shadowing more fundamental elements of a philosophy of language: speech and thoughts. They appear as logical instruments (although their utility in that case is never properly substantiated): but what does ‘corresponding’ amount to when the author has insisted that lekta are derivative on speech? The author’s repeated glosses of lekta as “the possibility of talking about impressions”, or that which is “required in order to speak through them” (p. 69) are ultimately unsuccessful at elucidating the relation.
Never making a distinction between a propositional content and a proposition, Alessandrelli gains little by his laborious effort to avoid the claim that all impressions have propositions as their contents. They are conceptual but not propositional, he says, in fear of committing himself to the view that the contents of impressions are incorporeal lekta — but what are concepts then? No answer is to be found in the book. He states that the content of an impression is not necessarily a specific kind of lekton arguing against a view put forward by Frede (‘The Stoic notion of a lekton’ in S. Everson, ed., Companions to Ancient Thought 3: Language, Cambridge 1994, 109-28) that rational impressions are rational precisely because they have a propositional content. This however does not imply that the content is a proposition. These are two rather different claims which are astonishingly confused in this book – a confusion which might have been dispelled with help from S. Bobzien, Die Stoische Modallogik (Wïurzburg 1986, ch. 1; there is a summary of some of the main points on this question found in her contribution to K. Algra et al., eds., The Cambridge History of Hellenistic Philosophy, Cambridge, 1999, 93-157), G. Striker, ‘Epicurus on the Truth of Sense Impressions’ (Archiv für Geschichte der Philosophie 59, 1977, 125-42) and J. Annas Hellenistic Philosophy of Mind (Berkeley, 1992) – all of which are absent from the references and argumentation in this book.
To settle his worry, maintaining the possibility of rational impressions with non-propositional contents, Alessandrelli opts for a distinction between sensorial and intellectual impressions — which indeed is mentioned in the texts, but which he however maps onto a distinction which is not mentioned as such in any text, namely between impressions which the author stipulates as causally generated from a sensible object and impressions which come about “in conformity with” external objects (as the author translates the epi with the dative), designating impressions whose objects are incorporeal lekta. True enough, there is a passage in Sextus Empiricus (M. 8.409) which raises the possibility of describing certain impressions (those not evidently caused by an external body actively impressing itself onto one’s sensorial receptors) in terms of “one’s soul being impressed as a result of them” (epi with the dative) — i.e. as a result of the external objects (not of lekta, note — although this process could also explain how we have impressions of lekta), and not “through their agency” (hupo with the genitive, used to designate the causal agent). Apart from this rather difficult and delicate passage, there are not, to my knowledge (nor, as it seems, to Alessandrelli’s), any further texts discussing of this point. The matter is one which, from its very wording, requires great caution in its analysis, as for example in J. Brunschwig, ‘La Théorie Stoïcienne du Genre Suprême et l’Ontologie Platonicienne’ (in J. Barnes and M. Mignucci, Matter and Metaphysics, Naples 1988, pp. 19-127 at 74-5), not mentioned on this point by Alessandrelli, or in D. N. Sedley, ‘Zeno’s Definition of Phantasia Kataleptike’ (in T. Scaltsas and A. S. Mason, eds., The Philosophy of Zeno, Larnaca 2002, 133-54). The latter gives a much more subtle reading in terms of “representational value” of what Alessandrelli over-simplifies into causally generated impressions for the sensorial kind.
There is, lastly, a great deal of freedom and trust put in our fragmentary and heterogeneous sources in this book. Sources are taken at face value which are in effect testimonies by writers who not only are not Stoic, but whose ties to Stoic doctrine and trustworthiness as faithful reporters are extremely thin. The last forty years of scholarship on these texts has at least taught us this much, that great caution and knowledge of context is crucial in using these texts to elucidate Stoic doctrine. For example, Alessandrelli builds his claims that lekta are semantic correlates of bodies with attributes, and that Cleanthes was the first to say so, on four lines in Clement of Alexandria’s Stromata (184.108.40.206), where he slips in the claim that “Cleanthes called predicateslekta, as did Archedemus”. A large claim placed on very frail if not crumbling shoulders. It is in the same pages that Alessandrelli, bolstering the image of Cleanthes as a logical innovator, attributes to Cleanthes the recourse to the Euthyphro mentioned above. However precious and rich in information this passage by Clement may be, it cannot constitute the one piece of evidence on which to build a whole thesis. The main result the reader gets from this claim specifically is that Cleanthes was not a particularly good logician, that in adding a derivative element, the lekton, to the semantics established by the Dialectic School and appropriated by Zeno, he complicated the previous notions of predicate and proposition but managed to explain very little. Instead, he merely creates problems, or indeed a problem: the one from this book’s title.
In conclusion, Alessandrelli has produced a book which puts centre-stage some crucial difficulties with Stoic logic and language. It is very focused on the instrumental role of lekta in language, which indeed ultimately is what lekta here are confined to. It is not at all interested in seeing the bigger picture or questioning the relevance of lekta to the whole Stoic system. It is not a book for beginners on this question, as it circumvents introductory questions and some fundamental readings; and for a more knowledgeable readership as well it might prove ultimately unsatisfactory.