Plato’s Philosophical Answer to the Three Deficiencies of the Written Word

In the Phaedrus Socrates gives a probably completely fabricated Egyptian story that relates the discovery of writing, as two mythological interlocutors differ in their appraisal of whether the new invention enhances or diminishes memory.  Socrates is clearly on the skeptical side, as he relates how Thamus probed the inventor Theuth on the utility of writing:

This will provide a forgetfulness in the souls of those who learn at the expense of memory, since they do not remind themselves by themselves internally, but because of a dependence on writing they are reminded externally by foreign impression  (translation mine, Phaedrus 275a2-5).[1]

Socrates adds, lest there be any doubt, that anyone who thinks that writing could instill anything clear and certain (saphes kai bebaion) would be full of simplicity (euetheias gemoi).  Socrates is just beginning to criticize writing, however, since he has two more accusations to level:

For writing certainly has this incredible feature and it is truly like painting.  The offspring of painting stand as living creatures do, but if you inquire anything, they are completely and reverently silent.  It is the same thing with words; you might think they speak on the grounds that they have some knowledge, but if you ask about any of the things spoken because you want to learn, [sc. ὁ λόγος] a word always signifies the one same thing alone.  And whenever it is once written, likewise the entire word rolls around promiscuously with those who understand, and with those for whom it is not at all fitting, and the word does not know for whom it is and is not necessary to speak to.  When it is wronged and reviled unjustly it always lacks the help of its father; for it is neither able to defend nor help itself  (translation mine, Phaedrus 275d4-e5).[2]

In summary, there are three main limitations to the written.  The first is that writing inhibits the cultivation of memory, making one dependent on orthographic conventions, onto which, in turn, our ideas are slavishly hitched.  Next, because written words are fixed and unalterable, they cannot clarify meaning or respond to questioning.  Lastly, and related to their inflexibility, words are unable to adapt to the needs of their audience.[3]  It should be noted that these same problems accrue to anything written, which is why Thomas Szelzak 1999: 31 has astutely commented that, “he [Plato] emphasizes the basic failings of writing, which are inherent in its nature.[4]  But whatever is inherent in a thing’s nature cannot be eliminated by a more or less skillful use of the thing.”

This raises the question, however, whether there is a counterpart to the written, a legitimate brother, possessed of the abilities which the written word lacks.[5]  This activity, in fact, is philosophy, and to be more precise, it is collection and division,[6] which provides the proper antidote to the problems of the written; for it is in collection and division that a method adequate to avoiding the perils of the static written word are found.


Now we will look at the benefits which collection and division possesses, by  emphasizing this process insofar as it compensates for, in Plato’s mind, a perceived set of deficiencies in the written word.[7]  The first is that in the process of collection and division there is a premium set on memory, for in a very Platonic sense, collection and division is quite straightforwardly a technical kind of remembering, (anamnesis) recollection:

For it is necessary for someone to comprehend what is said form by form, as it proceeds from many perceptions into one thing collected together by reasoning; and this is a recollection of those things which our soul saw then when it was traveling in procession with a god and scorned the things which we now say exist, and rose up into the real existence  (translation mine, Phaedrus 249b6-c1).[8]

We have here an explicit reference to collection and division.  As one proceeds to perceive the various particulars of whatever is the object of study, if one understands things “form by form,” then one simply is recollecting.[9]  It should be no surprise then, if, as this passage would lead us to believe, recollection is necessarily connected with the process of collection and division, that recollection is not spelled out in much detail.  Since collection and division, if it is successful, results in Platonic recollection and, as we have already noted, collection and division is a process which cannot be adequately explained in words.[10]  Whereas the serious accusation was leveled against written words that they do not help memory, but in fact are destructive of it, dialectic is here portrayed as that by which one is able to recollect the pure images of the forms.

So we have seen how collection and division has supplied the first defect found in the critique of writing: instead of weakening memory, collection and division is a work of memory par excellence.  Additionally, collection and division is also able to clarify and respond to an interlocutor, and unlike the second charge levied against writing, it is able to do so in a skillful way:

[The dialectician is] one who sees comprehensively divergent things in many places and leads them into one genus (ἰδέα), in order that by defining each thing he makes the thing clear concerning which on any occasion he wishes to teach about (translation mine, Phaedrus 265d3-5).[11]

Especially striking here, but which would ordinarily be missed unless we specifically had in mind the shortcomings of the written, is that collection, being spoken of here in isolation from division, is both clarificatory and responsive.[12]  It is clarificatory, in that while mere inscribed words, like a painting, signify the one same thing all the time (hen ti semainei mono tauton aei), collection is able to make a thing clear (delon).[13]  Yet this clarity is not reserved for the practitioner of dialectic alone, for he also is able, in teaching, to convey the nature of his knowledge to someone.[14]  Unlike that which is unresponsive and written, the practitioner of dialectic division is able to convey to his listener a kind of knowledge by making the object of his inquiry clear by defining each thing (ekaston horizomenos delon poie) in contrast to static written words which are entirely, solemnly mute (semnos panu siga).

The third and last judgment against the written is that it is unable to adjust to the demands of a proper audience.  Here again we see that dialectic, or collection and division, is able to compensate for this particular defect of written language.  In fact, some of this plasticity is due to the very nature of collection and division throughout the corpus, “Platonic dialectic is a method that is open; it does not develop through a specific plan. There is not a blueprint or a standard formula that is used by either Socrates or Plato”  (Kuperus 2007: 193).  Moreover, it is these very conditions, which in modern parlance we might say make it more of an art than a science, that, “dialectic is a skill to be acquired, much more than it is a body of propositions to be learnt”  (Robinson 1984: 74).[15]  If collection and division cannot adapt, it is no knowledge at all.[16]  Thus although it can be said that collection and division is not a method as set of directions, quite significantly it is a method insofar as one wishes to avoid the errors which vex the non-dialectician: poor memory, a lack of clarity and a non-adaptive “knowledge,” if it is even worthy of that qualified appellation.[17]



[1] τοῦτο γὰρ τῶν μαθόντων λήθην μὲν ἐν ψυχαῖς παρέξει μνήμης ἀμελετησίᾳ, ἅτε διὰ πίστιν γραφῆς ἔξωθεν ὑπ’ ἀλλοτρίων τύπων, οὐκ ἔνδοθεν αὐτοὺς ὑφ’ αὑτῶν ἀναμιμνῃσκομένους·

[2] δεινὸν γάρ που, ὦ Φαῖδρε, τοῦτ’ ἔχει γραφή, καὶ ὡς ἀληθῶς ὅμοιον ζωγραφίᾳ. καὶ γὰρ τὰ ἐκείνης ἔκγονα ἕστηκε μὲν ὡς ζῶντα, ἐὰν δ’ ἀνέρῃ τι, σεμνῶς πάνυ σιγᾷ. ταὐτὸν δὲ καὶ οἱ λόγοι· δόξαις μὲν ἂν ὥς τι φρονοῦντας αὐτοὺς λέγειν, ἐὰν δέ τι ἔρῃ τῶν λεγομένων βουλόμενος μαθεῖν, ἕν τι σημαίνει μόνον ταὐτὸν ἀεί. ὅταν δὲ ἅπαξ γραφῇ, κυλινδεῖται μὲν πανταχοῦ πᾶς λόγος ὁμοίως παρὰ τοῖς ἐπαΐουσιν, ὡς δ’ αὕτως παρ’ οἷς οὐδὲν προσήκει, καὶ οὐκ ἐπίσταται λέγειν οἷς δεῖ γε καὶ μή. πλημμελούμενος δὲ καὶ οὐκ ἐν δίκῃ λοιδορηθεὶς τοῦ πατρὸς ἀεὶ δεῖται βοηθοῦ· αὐτὸς γὰρ οὔτ’ ἀμύνασθαι οὔτε βοηθῆσαι δυνατὸς αὑτῷ.

[3] As Charles Griswold has pointed out, “Legitimate discourse is discovered by its speaker; it has as its primary goal self-instruction, and its secondary goal the generation of similar discourses in the souls of others (278a)” (Griswold 1986: 211).

[4] Pace Ronna Burger, who, in a transparent bid for special pleading, says, “Socrates’ critique of the silent written word is thus shown to be a condemnation of a part, and not the whole, of the art of writing.  The discriminating selectivity and power of self-protection which are denied to the illegitimate logos are, through that very denial, made manifest by the Platonic logos…”  (Burger 1980: 91).  However, it seems evident that the same objections stand just as firmly against Plato’s dialogues as they do any other writings.

[5] A masterful approach as to exactly how the Phaedrus as a whole can be taken as an exhortation to one philosophical life, among many, is given in Chapter 6, G.R.F. Ferrari (1987).

[6] Due to the self-imposed constraints on wordcount on this blog however, I will not go into details about collection and division here.

[7] Compare here the superiority of philosophy to rhetoric:  Ἐξαρκεῖ. εἰ γὰρ καὶ τοῦτό ἐστι διπλοῦν, τὸ μὲν ἕτερόν που τούτου κολακεία ἂν εἴη καὶ αἰσχρὰ δημηγορία, τὸ δ’ ἕτερον καλόν, τὸ παρασκευάζειν ὅπως ὡς βέλτισται ἔσονται τῶν πολιτῶν αἱ ψυχαί, καὶ διαμάχεσθαι λέγοντα τὰ βέλτιστα, εἴτε ἡδίω εἴτε ἀηδέστερα ἔσται τοῖς ἀκούουσιν.  ἀλλ’ οὐ πώποτε σὺ ταύτην εἶδες τὴν ῥητορικήν·  (Gorgias 503a5-9) “Good enough! For if this is also two-fold, of the two  one is certainly a flattery and a shameful public oratory, and the other noble, a preparative so that the souls of the citizens will be as good as possible, and it strives earnestly to say the best things, whether they be more pleasant to listeners, or more distasteful.  But you never yet saw this rhetoric.”

[8] δεῖ γὰρ ἄνθρωπον συνιέναι κατ’ εἶδος λεγόμενον, ἐκ πολλῶν ἰὸν αἰσθήσεων εἰς ἓν λογισμῷ συναιρούμενον· τοῦτο δ’ ἐστὶν ἀνάμνησις ἐκείνων ἅ ποτ’ εἶδεν ἡμῶν ἡ ψυχὴ συμπορευθεῖσα θεῷ καὶ ὑπεριδοῦσα ἃ νῦν εἶναί φαμεν, καὶ ἀνακύψασα εἰς τὸ ὂν ὄντως.

[9] I mean here by recollection the Platonic doctrine of remembering things in the here and now by dint of the soul’s previous association with them in a previous life.  The Phaedo has a particularly clear description: Καὶ μήν, ἔφη ὁ Κέβης ὑπολαβών, καὶ κατ’ ἐκεῖνόν γε τὸν λόγον, ὦ Σώκρατες, εἰ ἀληθής ἐστιν, ὃν σὺ εἴωθας θαμὰ λέγειν, ὅτι ἡμῖν ἡ μάθησις οὐκ ἄλλο τι ἢ ἀνάμνησις τυγχάνει οὖσα, καὶ κατὰ τοῦτον ἀνάγκη που ἡμᾶς ἐν προτέρῳ τινὶ χρόνῳ μεμαθηκέναι ἃ νῦν ἀναμιμνῃσκόμεθα (Phaedo 72e3-7). “Furthermore, according to that argument [of recollection], Socrates, which you are accustomed to make often, if it is a true one, then our learning happens to be nothing other than a recollection.  And in accordance with this, it is necessary that in some previous time we somehow learned those things, which we now recollect.”

[10] This is not to say that the theme of memory is not in the Phaedrus.  In fact, it is everywhere in the dialogue. Both μνάομαι and its cognates μνεία, ἀμνημονέω, ὑπόμνημα, including the just cited ἀνάμνησις (249c2) appear a total of 18 times.  It is interesting to note the effect, as a ὑπόμνημα, that writing is said to have on the individual is merely that of a reminding while the process of collection and division, as a ἀνάμνησις, is a remembering.  In Smyth this distinction is corroborated by ὑπό in composition meaning “slightly,” while ἀνά is “back” (1698.4, 1682.3)

[11] εἰς μίαν τε ἰδέαν συνορῶντα ἄγειν τὰ πολλαχῇ διεσπαρμένα, ἵνα ἕκαστον ὁριζόμενος δῆλον ποιῇ περὶ οὗ ἂν ἀεὶ διδάσκειν ἐθέλῃ.

[12] Division is described thus: τὸ πάλιν κατ’ εἴδη δύνασθαι διατέμνειν κατ’ ἄρθρα ᾗ πέφυκεν, καὶ μὴ ἐπιχειρεῖν καταγνύναι μέρος μηδέν, κακοῦ μαγείρου τρόπῳ χρώμενον· [Next, for someone] to be able to cut up again form by form according to the joints at which place it is natural, and not to attempt to destroy any part at all, making use of the manner of a bad butcher  (265e1-3).

[13] The descriptions quoted here of the written which contrast with division are from 275d4-e5.

[14] Charles Kahn 1999: 372 has noted that, “[After Socrates’ palinode in the Phaedrus] Plato argues that philosophical dialectic, the systematic study of unity and plurality, provides the foundation for all rational inquiry and all successful discourse.”

[15] Lewis Campbell 1867: xi has artfully expressed the exotic epistemology behind Plato’s doctrine of the relationship between the mind and the written word:  “Plato never conceived…that a new method could possibly level intellects, or become a substitute for invention.  He never imagines a form of thinking as separable from thought.”

[16] Miles Burnyeat 2012: 187 explains the importance of this idea, “”It is a direct consequence of this epistemological stance [i.e. knowledge comes through the mind] that there is no such thing as historical knowledge or knowledge transmitted by the word of another person.”

[17] Method (μέθοδος) is not so distinct and stepwise a process as is normally imagined.  I agree with the sentiment of Paul Woodruff 2007: 153, “When I speak of method, I do not mean it in the sense of a modern, scientific method, but in the original Greek sense of ‘being on or along a road or pathway.’”


Burger, Ronna. “The Art of Writing.” Plato’s Phaedrus: A Defense of a Philosophic Art of Writing. University: University of Alabama, 1980. 91.

Burnyeat, Myles. “The Passion of Reason in Plato’s Phaedrus.” Explorations in Ancient and Modern Philosophy. Cambridge: Cambridge UP, 2011. 238-58.

Campbell, Lewis and Plato. Introduction. The Sophistes and Politicus of Plato,. Oxford: Clarendon, 1867. Xi.

Cooper, John M. Ed., and D. S. Hutchinson, eds. Plato: Complete Works. Indianapolis: Hackett Publishing, 1997.

Ferrari, G. R. F. Listening to the Cicadas: A Study of Plato’s Phaedrus. Cambridge: Cambridge UP, 1987.

Griswold, Charles L. “Theuth, Thamus, the Criticism of Writing, and the Praise of Dialectic.” Self-knowledge in Plato’s Phaedrus. New Haven: Yale UP, 1986. 203-18.

Kahn, Charles H. “12 Phaedrus and the Limits of Writing.” Plato and the Socratic Dialogue. Cambridge: Cambridge UP, 1999. 372.

Kuperus, Gerard. “Traveling with Socrates: Dialectic in the Phaedo and Protagoras.” Philosophy in Dialogue: Plato’s Many Devices. Evanston, IL: Northwestern UP, 2007.

Liddell, Henry George, Robert Scott, and Henry Stuart Jones. A Greek-English Lexicon: With a Supplement 1968. Oxford: Clarendon, 1983.

Piccone, Enrique. “Four Features of Dialectic in Plato’s Phaedrus.” Understanding the Phaedrus: Proceedings of the II Symposium Platonicum. Sankt Augustin: Academia Verlag, 1992. 261-64.

Plato, and Ioannes Burnet. Tetralogias III – IV Continens. Oxonii: Clarendon, 1984.

Plato, and Rowe. C.J.. Phaedrus. Warminster, Wiltshire, England: Aris & Phillips, 1986.

Robinson, Richard. “VI Dialectic.” Plato’s Earlier Dialectic. Oxford, Clarendon, 1984.

Ross, W. D. Plato’s Theory of Ideas. Oxford: Clarendon, 1951.

Sayre, Kenneth. “A Maieutic View of Five Late Dialogues.” Oxford Studies in Ancient Philosophy. Oxford: Clarendon, 1992. 221-244.

Smyth, Herbert Weir, and Gordon M. Messing. Greek Grammar. Cambridge: Harvard UP, 1956.

Szlezák, Thomas Alexander. “The Critique of Writing in the Phaedrus.” Reading Plato. London: Routledge, 1999. 31.

Werner, Daniel. “Plato’s Phaedrus and the Problem of Unity.” Oxford Studies in Ancient Philosophy. Vol. 32: Summer 2007. By David Sedley. Oxford: Oxford UP, 2007.

Wohl, Victoria. “Chapter 2: Pornos of the People.” Love among the Ruins: The Erotics of Democracy in Classical Athens. Princeton, NJ: Princeton UP, 2002.

Woodruff, Martha K. “Plato’s Different Device: Reconciling the One and the Many in the Philebus.” Philosophy in Dialogue: Plato’s Many Devices. Evanston, IL: Northwestern UP, 2007









Problems with Plato: Animal Diversity and Robust Division

Yesterday, I brought up some difficulties that occur in the first method of division presented in Aristotle’s Parts of Animals.

“Some construe the individual species by dividing the group into two differentiae.  In one way this is not easy, in another impossible.  For of some there will be only one differentia, but the other terms will be superfluous, such as in the case of footed, two-footed, footed with parted toes.  For this last differentia alone is proper” (Parts of Animals, 642b5-9)?

Before discussing the difficulties again, it must be taken into consideration, of course, that since he is embarking on a criticism of Academic methodology, any philosophical difficulties arising from this process of division might in fact owe to chinks in the armor which Aristotle himself was trying to illuminate.  He might very well be bringing some of these difficulties to light without proceeding to explain them.

A couple of the difficulties I brought up yesterday were:

“It is not true that the concept of “footed with parted toes” includes “two footed” and “footed.” viz. lizards

It is also clear from his use of the term elsewhere, that “footed with parted toes (σχιζόπους)” is not meant to designate any particular species alone, such as humans (cf. HA 593a28, concerning birds).”

A mistake one could make, at least in Aristotle’s construal of the division, is failing to recognize that at every point of the division a particular animal is being guided through each step of the division.  So, in this example, one has to have in mind a particular animal which one leads through each “gate” of the division.  There is no abstracted “two-footed,” in other words; differentiae always belong to real animals.  (This is one reason why Aristotle says shortly that there cannot be divisions of non-being, for no animals correspond to such a division.)

In fact, Aristotle will return to this division at 643b29 ff.  He says there two interesting things.  He affirms that with this method both (a) that only one differentia will be arrived at (b) it is impossible for one differentia to be adequate for a species.  In light of this he brings up the same “footed, two-footed, footed with parted toes,” tri-partite division.  Only this time he applies it to “man” (ἄνθρωπος).  He points out that man is many other things besides possessing parted toes.

Both of these judgments merge into a single criticism: there is a need for there to be multiple differentia, although the Academic method cannot allow for it, it is not so much wrong, as it is inadequate.  One reason for a need for more lines of division is that there are many homologous features in animals.  Similar body plans, parts, and functions means that there will be many animals with split-toed feet.  Because of this diversity there will be a corresponding need for a robust method of division.