Introduction to the Greek Apophthegmata

The SAWS Project offers users diplomatic and normalized transcriptions of fourteen related Greek collections of wise and witty sayings (gnomologia), some of which have never before been published, as well as a preliminary critical edition with apparatus, translation and references. The gnomological collections edited here are alphabetically arranged in the sixteen manuscripts; the best known is the Gnomologium Vaticanum from Codex Vaticanus Graecus 743. Users are advised to read through the following sections in order to understand and use the material correctly. For updated information concerning the SAWS Greek gnomologia, please visit Ars Edendi.


Anthology We use the terms anthology and florilegium interchangeably to denote collections of quotations in both verse and prose from one or more authors. An anthology or florilegium may sometimes contain anecdotes and sayings attributed to individual speakers.

Apophthegm The Latin is apophthegma from the Greek ἀπόφθεγμα (from ἀποφθέγγεσθαι). The terms “apophthegm” and “chreia” are used here to describe the same kind of saying, that is, a very brief narrative whose purpose is to convey the words or actions of some well known personage, the briefest form being “so-and-so said this”. A very typical form would be, “When told or asked that, so-and-so answered this”. Unlike Chreia and Gnome, apophthegm is not a technical term in rhetoric. Diogenes Laertius uses it frequently to characterize the sayings of the philosophers that he records. It is also the usual term used in the lemmata of mss. to designate the kind of sayings technically known as chreiae as distinct from gnome. There are well known collections of apophthegms from antiquity, including the apophthegms of Spartans (cf. Λακωνικὰ ἀποφθέγματα in Aristot. Rhet. 1412a 23), and collections like Apophthegmata regum et imperatorum transmitted under Plutarch’s name. (A somewhat similar term is apomnemoneuma, best known from the Greek title of Xenophon’s Memorabilia, but we do not use this term within SAWS).

Chreia This term is used in the present work as equivalent to Apophthegm. The ordinary Greek word χρεία (sometimes transliterated by other scholars as chria) received a technical meaning as designating a certain form of saying at some point in the Hellenistic period. A typical definition is that found in Aphthonius (Progymnasmata, Rhet. gr. 10 p. 3): Χρεία ἐστὶν ἀπομνημόνευμα σύντομον εὐστόχως ἐπί τι πρόσωπον ἀναφέρουσα (“A chreia is a concise Apomnemoneuma aptly referred to some person”). We may suppose that the term originated from some such expression as ἀποφθέγματα χρειώδη (cf. Diogenes Laertius 4.47, 5.39).

Florilegium We use this as equivalent to Anthology. A sacro-profane florilegium denotes a collection of quotations (and/or sayings) from both Judaeo-Christian and pagan Greek and/or Roman texts.

Gnome The English word comes from Greek γνώμη; a synomym is maxim. The Latin equivalent is sententia. As a term for a certain kind of saying, gnome is defined already by Aristotle as a universal proposition regarding human actions (cf. Rhet. 1394a 21 ff.), a definition more or less upheld and developed by later authors of exercise books in rhetoric (Progymnasmata). The term is used similarly in this edition to denote pithy expressions of principles having to do with ethical matters. Similar terms such as “aphorism” or “axiom” are used here to denote pithy statements of (non-ethical) philosophical and scientific principles. Maxims or gnomes are anonymous in themselves but are usually attributed to specific authors by means of a lemma.

Gnomology This is the English equivalent of the Latin gnomologium. The Greek word γνωμολογία is used by classical authors to denote gnomic discourse or the sententious style (cf. Plat. Phdr. 267C, Aristot. Rhet. 1394a ff.). Its use as denoting a collection of sayings seems to be medieval at the very earliest. The word gnomologium or gnomology is used here to denote a collection consisting primarily of maxims (gnomes) or of apophthegms relating to one or more authors. It may be arranged either alphabetically or thematically or by author. Thematic arrangement is less frequent.

Homoeoma This transcribes the Greek ὁμοίωμα “likeness” in the sense of simile or similitude. It refers to a gnomic saying in the form of a similitude normally accompanied by a brief explanation of the image. Example: “The life of misers is like a dead man’s supper: it’s got everything but someone to enjoy himself”.

Maxim Used as a synonym for Gnome.

Proverb This or the Greek παροιμία will be used in the usual sense of a pithy anonymous saying in common and recognized use.

Saying This is, simply, a catch-all term here for any kind of saying (gnomic, anecdotal, proverbial, aphoristic, etc.).

Greek Gnomological Traditions

Wisdom sayings form a familiar part of speech communities throughout the world, and 19th century philologists as well as contemporary literary scholars and socio-linguists have studied them in some depth. Ancient Greek thinkers distinguished between various types of  sayings, some of which are to be found in our terminology section. For our purposes the two most important sayings types are the gnome and the apophthegm. Other textual items occur in the collections edited here, such as brief fables or short quotations of literary works, but the vast majority fall into the category of gnome or of apophthegm, with the latter greatly predominanting.

Two basic gnomological traditions must be distinguished: one is the traditional use of gnomes in poetry and prose from early on; the other is the systematic collection of gnomes in various kinds of gnomologia (gnomologies). The gnomic mode of expression is found in abundance from Homer onwards in both poetry and prose; the corpus of gnomic verse attributed to Theognis is our first extant collection. The effective use of gnomes, pithily expressed universal statements, is a staple of classical rhetoric and is specifically addressed in Aristotle, Rhetoric II.21 (1394a 19 – 1395b 20). The earliest prose collection of gnomes is probably the sayings of the legendary Greek sages (the seven wise men) were said to be inscribed at Delphi (cf. Plato, Protagoras 342e-343b); Hellenistic inscriptions with these sayings have been found, and a collection of them is ascribed to Demetrius of Phaleron (cf. Stobaeus 3.1.172). The sayings of the seven sages are continuously transmitted in anthologies and gnomologies from ancient to medieval times.

Compositions based on the "gnomes of the poets" probably already formed a part of school exercises in classical times. Hellenistic and later rhetoricians introduced exercises using the more anecdotal but still brief form of a saying known as the chreia for which we use apophthegm as a synonym.

Diogenes Laertius cites a number of Hellenistic collections of chreiae, although none of these are extant. Our first extant collections are ascribed to Plutarch in the first century of our age. Gnomes and chreiae formed two separate categories in the school exercises and, it seems, they were kept separate in the collections at the start. The first extant anthology that combines them under the same thematic headings is that of Stobaeus from the fifth century; though we may assume they had already been combined in previous collections no longer extant. Stobaeus was a major source for medieval collections, but the medieval compilers had also access to other late antique sources. It was not until the tenth century that Christian and non-Christian sayings were combined under the same thematic headings, e.g. in the widespread florilegium of pseudo-Maximus (Loci communes or commonplaces). Examples of these and other thematic headings are presented here, Greek florilegia: Pinakes.

Apophthegms (chreiae) are particularly associated with Diogenes and the other Cynic philosophers; the apophthegmatic style was a peculiar feature of Cynic discourse. However other philosophical schools influenced the gnomological tradition, e.g. collections of sayings are attributed to some of the peripatetics such as Demetrius of Phaleron. Significant collections of maxims and sayings of Epicurus, Democritus, Pythagoras and others are an important part of the gnomological tradition. Furthermore, gnomic sayings continued to be culled from poetic sources in post-classical times; those attributed to Euripides and Menander are in great abundance.

Collections of sayings are normally arranged by (1) author, (2) theme, (3) alphabet. A collection arranged by author may contain only a single author (Euripides, Menander, Epicurus). If a collection contains more than one author, the authors may be arranged in various ways, e.g. chronologically and geographically as in Plutarch's Sayings of Kings and Commanders, or alphabetically as in the collections edited here. As time went by, variously arranged collections were combined and rearranged, so that by the time we reach the Middle Ages a glorious confusion generally reigns, though careful philological work can at times unravel some of the various strands and trace the medieval collections back to earlier sources. However, we must also allow for the creation of new sayings or at least rewriting of received sayings throughout the entire period, given that gnomes and apophthegms were a staple feature in school instruction, and students were taught to manipulate them in public speaking and in composition.

In 1882 Curt Wachsmuth published a gnomological collection from the 13th century codex Vindobonesis theologicus 149 which he called Die Wiener Apophthegmen-Sammlung and which we refer to as WA. The codex otherwise chiefly contains commentaries on Gregory of Nazianz. The apophthegmata in WA are attributed to various Greek and some Roman celebrities arranged in one-letter deep alphabetical order by personal names. The collection is incomplete in the manuscript, lacking selections for letters B-Π. In spite of the fact that the collection is already greatly reduced in the manuscript, Wachsmuth still omitted a couple of pages of sayings in his edition, thinking they did not belong to the original collection, although all of them, with one exception, later turned out to be in Gnomologium Vaticanum.

Gnomologium Vaticanum or GV is the name of a collection of 577 apophthegms, arranged in WA, in the 15th century codex Vaticanus graecus 743 published by Leo Sternbach five years after Wachsmuth’s publication of WA. Unlike the Viennese manuscript, Vatican 743 is dedicated exclusively to gnomological material. GV covers the complete alphabet; a collection of sayings of famous women is added at the end, as in WA. The contents, sequence and wording in the two collections are close, though not identical. WA contains, for example, some gnomic verses absent in GV. A quotation from a late antique Aristotelian commentator has also been included under the name of Aristotle in WA. Other collections related to WA and GV have subsequently been discovered. All of these share the alphabetic arrangement, have mostly the same sayings (allowing for individual additions and omissions) and usually more or less the same wording. We have given the whole tradition the collective title Apophthegmata et Gnomae Secundum Alphabetum (AGSA), which is based on the title found in WA.

Most of the persons cited in AGSA are Greek philosophers, although there are also a large number of apophthegms belonging to other famous authors and artists (tragic and comic poets, orators, historians, kings, generals, etc.), including even Cicero, who is the latest identifiable name. It is notable, indeed, that some highly quotable authors, whose sayings circulate in other collections, such as Epicurus, Epictetus, Isocrates, Democritus, are only minimally present in AGSA. Moreover, authors belonging to the Imperial period are nearly all missing, though there is the notable exception of Demonax, a second century “Cynic”.

We will now look at the so-called Corpus Parisinum (CP), the contents of which will serve as a summary of the gnomological traditions relevant to the Greek editions of the SAWS project. CP is primarily represented by codd. Parisinus graecus 1168 and Bodleianus Digby 6 (P and D). Ms P contains only the corpus while D has a few other texts at the end. CP is a collection of collections. It was the main source for the pagan selections in a large florilegium that we call the Loci Communes of pseudo-Maximus Confessor. The oldest manuscript of this florilegium is from the late 10th century, so CP must have existed in some form already earlier in the 10th or in the 9th century, a time when a number of manuscripts were being produced that were systematic compilations of ancient Greek texts. Part of it is arranged alphabetically as in AGSA and it clearly used a collection related to AGSA as one of its several sources. This means that the AGSA collection must have existed prior to CP, so this moves us back centuries before the actual manuscripts containing AGSA.

Mss P and D share the following collections in the same order and arrangement:

CP 1 (568 selections), the single largest part of the collection, consisting of quotations taken from the Christian Fathers, the wisdom literature of the Old Testament and, in a few cases, from the New Testament, as well as from the Jewish philosopher Philo. It is an anthology arranged by author’s name (or title of biblical book), although in no apparent order. The major source for CP 1 is the anthology known as Sacra Parallella attributed to John of Damascus.

CP 2 (14 selections), a series of theosophic oracles foretelling Christian dogmas but attributed to pagan Greeks.

CP 3 (556 selections), the second largest section of CP. It contains gnomes, apophthegms and a number of brief quotations (e.g. from Greek novels of late antiquity) associated with eighty names, all but two of them pagan. It can be characterized as a gnomology arranged by author, although, again, in no immediately apparent order. Within each authorial section, there is a fairly systematic attempt to keep gnomes and apophthegms separate. This is the section for which CP is most famous. Among other things, it contains a long series of ethical sayings attributed to Democritus.

CP 4 (214 selections), extracts from the florilegium of Stobaeus. Unlike the preceding section, these excerpts are arranged thematically with approximately the same headings as in the chapters of Stobaeus from which they mostly are taken. Hense made use of Anton Elter’s transcription of ms. P for his edition of Stobaeus.

CP 5 (97 selections), an abbreviated version of a collection of maxims from the gnomology variously called Democritus, Isocrates and Epictetus (DEI or DIE) also known as Gnomologium Byzantinum, a gnomology first published by Curt Wachsmuth. Like the other extant versions of the same gnomology, CP 5 arranges these maxims in short thematic chapters.           

CP 4 B (16 selections), a brief series of maxims and apophthegms arranged thematically as in the two preceding sections. All of the selections derive from Stobaeus and have ended up here no doubt due to some displacement in the course of transmission. They really belong to CP 4. They have been left in the edition in the position in which they occur in P and D but given the number of the part to which they belong.

CP 6 (228 selections), a substantial collection of apophthegms related to AGSA. Thus, CP 6 is alphabetically arranged primarily by author, although a number of anonymous gnomic sayings are added in alphabetical sequence according to initial letter. These sayings stem from a source related to the ΑΠΜ or APM (Ἄριστον καὶ Πρῶτον Μάθημα).

CP 7 (304 selections), an alphabetically arranged series of monostichoi attributed to Menander. As we can see from the Loci communes of pseudo-Maximus, this section did not form part of the original CP.

For the purposes of this introduction, we can ignore sections 1-2 and 7 above. The compiler(s) behind the rest of CP did not simply bring together representative collections from each separate tradition. The third section above, CP3, draws from the traditions represented in CP 4, 5 and 6, i.e. from Stobaeus (and apparently Diogenes Laertius), AGSA, DIE, ΑΠΜ and other sources. For whatever reason, the compiler(s) of CP3 did not complete an integrated edition of all these traditions, though that was apparently the purpose of CP3. Instead we find an attempt at an integrated edition of gnomological sources arranged by author in CP 3, with leftover materials left in the original arrangemnet of the source collections in CP 4, 5 and 6.

This brief survey of CP shows how complicated the gnomological tradition can be and provides the reader with indications of the most relevant gnomological traditions that receive mention in our edition.

Sources of SAWS Greek Gnomologia

Here follows an alphabetic listing of the primary sources used for our edition. Manuscript dates are for the most part based on the information in the corresponding manuscript catalogues.

Appendix Gnomica: Leo Sternbach edited a collection of 123 sayings, mostly apophthegms, from three mss. and called it Appendix Gnomica. The codices are: Laurentianus graecus 86.8 (ff. 314r-316r) = Laur; Parisinus supplementum graecum 690 (ff. 18v-19v) = Paris; Vaticanus graecus 742 (ff. 63v-70r) = Vat. We have recollated these three manuscript copies for our edition. For Par. suppl. gr. 690 we used a microfilm copy that was rather poor; we did not have the time to acquire a better copy or inspect the original.

Appendix Vaticana I (AVI): This is the first of two series of sayings found in cod. Vaticanus Graecus 1144 (ff. 215v-225v) edited by Sternbach. He published this series of 371 sayings as though it were a single collection. The first 120 sayings, however, belong to the AGSA-tradition, whereas nos. 231-371 represent one of the main sources for Elter’s Gnomica Homoeomata (Class R). The sayings in-between, nos. 121-230, are a strange mixture, the first fifteen or so of them having parallels in DIE, and ΑΠΜ, among others. The remainder seem to be quotations that stem from, among other sources, collections of scholia and similar kinds of reference works. We have done a new collation for our edition.

Appendix Vaticana II (AVII): This is the second series of sayings found on ff. 228r-232v of the same ms. as AVI (Vat. gr. 1144). It consists of 147 sayings, all but a handful of which clearly stem from the AGSA collection. We have done a new collation for our edition.

Corpus Parisinum VI (CPVI): The two witnesses are Parisinus graecus 1168 or P (Paris), and Bodleianus Digby 6 or D (Digby). Our edition is based on Searby 2007, but also represents a new collation. CPVI is the section most obviously related to AGSA, even though CPIII also drew from the same tradition. See the survey of CP above.

Florilegium Leidense (FL): Rynhard Beynen published this from cod. Voss. gr. Q 13 FL 2r–15r, 16th c., is practically identical with FM, which see. FL was used by Wachsmuth for his Gnomologium Byzantinum (DEI = Democritus Epictetus Isocrates). We have done a new collation for our edition.

Florilegium Monacense (FM): This collection from cod. Mon. gr. 8, FM 39r–45v,16th c. was first published by C. Walz, republished later by Meineke. (Walz adds no notes to his edition, so it is often difficult to distinguish between his corrections and his own possible misreadings.)  FM and FL are independent copies of the same source. FM and FL stem from three collections transmitted together: (1) FM 1-101 (FL 1-97), a recension of DEI used by Wachsmuth for his Gnomologium Byzantinum; (2) FM 102-154 (FL 98-144), poetic γνῶμαι stemming from the anthology of Orion; (3) FM 155-270 (FL 145-257), a witness to AGSA. We have done a new collation for our edition.

Gnomologium Vaticanum (GV): This is a collection of 577 apophthegms, alphabetically arranged by name of author, in cod. Vat. gr. 743 (ff. 6-46v), 15th-16th century, edited by Leo Sternbach . It is the largest and best known representative of the AGSA tradition. We have done a new collation for our edition.

Florilegium Chisianum (FC): FC is a previously unedited collection of 109 apophthegms found in cod. Chis. Gr. R IV 11 (13th c.), ff. 105r–112v. The sayings are arranged alphabetically by author, covering the letters A-P. FC contains all apophthegms found in FL and FM for these letters as well as several others. We have not yet concluded if FC is best understood as an expanded version of (the source of) FL/FM, or (the source of) FL/FM as a reduced version of FC. Unfortunately our images of this manuscript were somewhat deficient and some readings remain uncertain until we check them in the original or acquire better images.

Florilegium Vindobonense (FV): FV is a previously unedited collection of 316 sayings found in cod. Vind. Gr. Phil. 154 (14th c.), ff. 369v–375v. It can be divided into four main parts: (1) FV 1-66, a collection of apophthegms with many parallels in AGSA; (2) FV 67-212, a collection of gnomes almost identical to section 1 and 2 in FL/FM (FL 1-144/FM 1-154); (3) FV 213-311, a collection of apophthegms almost identical to section 3 in FL/FM (FL 145-257/FM 155-270); (4) FV 312-316, five sayings attributed to Plutarch and Basil the Great. We have not yet reached clear conclusions about the exact relationship between FV and FL/FM.

Florilegium Baroccianum (FB): FB is a previously unedited collection of 124 sayings found in cod. Bar. Gr. 111 (15th c.), ff. 58r-62r. The collection is an abbreviated version of FV drawing material from all four parts of FV.

Florilegium Vaticanum (FVat): FVat is a previously unedited collection of 46 sayings found in cod. Vat. gr. 872 (13th c.), ff. 244ar-247r. The first 44 sayings make up a slightly abbreviated version of the first part of FV. The last two sayings have a different origin.

Florilegium Neapolitanum (FN): FN is a previously unedited collection of 19 sayings found in cod. Neap. gr. II E 5 (14th c.), f. 1r. It is identical to the ending of FL/FM (FL 239-252/FM 257-270).

Florilegium Parisinum (FP): FP is a previously unedited collection of 35 sayings found in cod. Par. Suppl. Gr. 690 (12th c.), ff. 28v-29r, a manuscript which also contains one of the versions of ApG. FP mainly consists of apophthegms attributed to Alexander, but some other characters are represented as well. These apophthegms have many parallels in AGSA, but we have not yet fully investigated the nature of FVs relation to AGSA. It should also be mentioned that the manuscript contains some additional apophthegmatical material and excerpts related to Alexander.

Die Wiener Apophthegmen-Sammlung (WA): This was edited by Curt Wachsmuth in 1882 from cod. Vindobonensis theol. 149, ff 302v-308r. It is incomplete, lacking selections for letters B-Π. Ff. 306r -307 v contain sayings omitted by Wachsmuth in his edition, following the arguments of Diels (1874), who held that they did not belong to the original collection. We have done a new collation of the manuscript and include these sayings.


Guide to SAWS Greek Gnomologia

Users are advised to acquaint themselves with the introductory sections on terminology, traditions and sources, before turning to the editions themselves.

The recommended point of entry into the Greek texts is through the Edition. Like the underlying sources, the edition is arranged by author; all the sayings attributed to a given author from all the sources are placed together. There are a number of multiple attributions of basically the same saying; to deal with this, we edit the version attributed to the author with the best basis in the sources and link to this under the names of the other attributed authors.

By opening a given saying in the edition, users can view the critical apparatus and an English translation for the selected item. References to parallels in other sources and some bibliographical references are provided. 

Each of the edited collections can also be viewed on its own in a diplomatic as well as a normalized transcription.  In this way users can see the individual edited items, the wider tradition of which may go back to Hellenistic or even Classical times, as well as the context in the medieval manuscript tradition.

We are not trying to reconstruct an archetype of a putative original AGSA collection (Apophthegmata et Gnomae Secundum Alphabetum), the reason being that we do not yet see how such a reconstruction is possible. Instead we include all the material present in the manuscript sources that have been identified.

The main principles of the edition are as follows:

(1) Each edited item is introduced by its own number and a list of witnesses to that item.

(2) The witnesses are listed in approximate chronological order according to manuscript date; in the case of a source based on manuscripts from different periods (e.g. Appendix Gnomica), we list it according to the oldest manuscript. Manuscript dating is normally based on information provided in manuscript catalogues.

(3) The version on which the edition of an item is based is regularly taken from the oldest manuscript; exceptions are sometimes made out of convenience, e.g. when selecting another manuscript source makes for a more compact apparatus. The base version is printed in boldface.

(4) In the critical apparatus, we do not show insignificant orthographical deviations or other simple errors, all of which can be seed by going directly to the diplomatic view of each source.

(5) We try to avoid making corrections to the base version, apart from normalization; sometimes we do correct the text in the base manuscript using the readings in the other manuscripts, when it seems to us that no real sense can be made of the base version without relying on the other sources. This principle means that we often settle for unclassical usage or a less smooth text than would be otherwise possible.

(6) We do not use external sources to emend the texts, although we may be guided by parallels, for example in Diogenes Laertius, in our choice of which version to privilege in the edition of an item.

This edition of Greek gnomologia remains in a preliminary state: it is still a work in progress. To date we have not been able to add all the references we have at our disposal or which we intend to locate; we have not yet finished adding corrections. Furthermore there remain inconsistencies in editorial method and selection of material which need to be worked out in the future. Updated information about improvements to the material will be posted by the responsible editors on the homepage of the Ars Edendi project at Stockholm University (Ars Edendi). Users are encouraged to report any errors or suggestions about how to improve usability to Denis Searby (denis.searby [at]