НИКОЛА БЕНИН
DURING THE LAST HUNDRED YEARS research on symbols and myths has greatly
expanded and has produced authentic insights, especially in the general
science of signs, called semiotics, and in linguistics, psychology, and anthropology. It would appear worthwhile to ask if and how the emerging new
methods, in particular those of the structuralists, can be related to the study
of medieval symbolism. The first part of this paper will address the question
in a general way; the second and larger section will deal with one particular
group of symbols.
I
The Greek noun "symbolon" is derived from the verb "symballein," meaning "to throw together, bring together, put together," also "to collect" and "to
compare."
It is important to realize that in classical Greek "symbolon" was at first
literally related to a "drawing together." In a meeting or party the symbols
could be contributions to a shared meal.1 In the control mechanisms of a
contract, the two parts of a token, which were to be separated by the
contracting parties and brought together again, were called symboland had
the function of tallies.2 In both cases there is an element of contrast (there
are at least two "partners" in a party or in a contract) and an element of
likeness (the "partners" share a purpose or the "parts" fit together). Thus
even the earliest terminology implied that the symbol is different in some
ways from that which it represents and nevertheless in some ways similar or
* This paper is a revised and annotated version of the presidential address delivered on April
14, 1978, at the Annual Meeting of the Mediaeval Academy of America, held at Yale University.
1 Liddell and Scott, Greek-English Dictionary, s.v. a6gP4okov I, 3, also s.v. cwgp4oxi III.
2 Ibid., s.v. a6j43okov I, 2. For a fuller history of the Greek term to the fourth century A.D.,
see W. MuIri, ?YMBOAON: Wort- und sachgeschichtliche Studie, Beilage zum Jahresbericht uber
das Stadtische Gymnasium in Bern (Bern, 1931); cf. also M. Schlesinger, Geschichte des Symbols
(Berlin, 1912; repr. Hildesheim, 1967), pp. 5 ff.
223
224 Understanding of Symbolism
even one with it.3 These are general traits of symbolism to which I shall have
to return more than once.
All the various meanings that we still attach to the term symbol go back to
pagan or Christian antiquity. The Fathers of the Church used the term in
two senses. They spoke of a Symbol of the Christian Faith, by which they
meant the Creed. It was a "drawing together," a summary and token, of the
main truths and doctrines of Christianity.4 This remained the principal
meaning of symbol in the medieval West; it is still important in Christian
theology and liturgy, above all in the so-called Symbols or Creeds of the
Apostles and of the Councils of Nicaea and Constantinople.5
The second meaning of symbol that was current in early Christian times
has much wider applications, and it is the one that is chiefly relevant for our
topic. Here symbol is very close to, or even synonymous with, semeion, signum,
sign - Origen says that whatever happens in an unexpected or strange way
in Holy Scripture is a "semeion kai symbolon," a sign or symbol, of something else, namely of something beyond the realm of sense experience.6
"Symbolon" as a term that means a sign with deeply spiritual and even
mystical meaning appears full-fledged in the Greek Christian world around
the year 500, in the works of the great unknown who calls himself Dionysius
the Areopagite. The title of the second chapter of his treatise On the Heavenly
Hierarchy announces its contents by stating that the divine and heavenly
fittingly appear also in symbols which have no obvious likeness to those
exalted realms.7 The meaning of the whole chapter is that since God is so
high above human conception, it may be more revealing to express the
divine and heavenly by phenomena taken from the lower reaches of the
created cosmos than it is to choose symbols that superficially seem closer to
God. Thus, using biblical symbolism, he says that not only the light of the
sun or the stars, but also a wild animal, such as a lion, or a stone rejected by a
builder, may be symbols of Christ.8
3 For the distinctness of similarity and unity, and related distinctions of analogy from participation and metaphor from metonymy and synecdoche, see below, passim.
4 For symbolon as password, formula of belief, and sacrament in the non-Christian mystery
cults of antiquity, see Muri, op. cit., pp. 39 ff.
5 See J. de Ghellinck, Les recherches sur les origines du symbole des ap6tres = Patristiquet moyen dge,
1 (Gembloux, Bruxelles, Paris, 1949).
6 Origen, InJoann. 13.60, PG 14:521: o6ic ?aTI Tt icapd6o4ov yev6gevov b Tf Ipawpi, B tin ?atI
crl0tov i(at Ct3o0k0ov ?t?poI) iap6 T6 aia0iTb5 y8y8vnIvov.
7 Ps.-Dionysius, De coelesti hierarchia, cap. 2, PG 3:136: 'Ott irpvr6vTOq Ta 0?ta Kai oip&vta icai
6ta Tdv 6vojioiov carutp6kov &1cpaiv8tw.
8 Ibid., p. 144 C ff.: Kai icoT? gtv ai)Tiv (i.e., divine government) &ir6 Tb patvogvOv Ttijov
igvorkntv, bvO fjklov 6tratocnfvi (Malach. 4.2), bS 'acTpa T6v ?&4ov (Apoc. 22.16) ... noTt 6& &i6n
T6V ?a TOV . . . bS Xi0ov ?arLpoyovlatov (Ephes. 2.20). 'Aa& Kcai 0nptogioppiav a6tf
irpiTi0?acnt, ical Xo0VTog a)TD ical ir6v0rpog i66t6lTa nept6nTmtooc (Osee 13.7) . . . (145A) O66?v
O?V &TOlOV, ?i ical T6q OUipavia' o6OiaO 'Cy T6)V &6RgPaiVO1)a6V &vogioiov 6jioioTfToV &vwRX6TToai. . . . This kind of argumentation is symptomatic for the influence of Proclus on the
Pseudo-Dionysius. See R. Roques, L'univers dionysien (Paris, 1954), p. 115, J. A. Coulter, The
Literary Microcosm: Theories of Interpretation of the Later Neoplatonists (Leiden, 1976), p. 50; J.
Pepin, "Aspects theoriques du symbolisme dans la tradition dionysienne," Simboli e simbologia
Understanding of Symbolism 225
I shall cite only one Latin commentator on the Areopagite, from the
twelfth century, Hugh of St. Victor.9 A text from Hugh's commentary on the
Heavenly Hierarchy is quite characteristic of the medieval understanding of
symbolism: "A symbol is a collecting of visible forms for the demonstration
of invisible things."'10 "Collecting" translates collatio, which itself is an almost
literal translation of the Greek symballein.
In the examples from Origen, Pseudo-Dionysius, and Hugh of St. Victor,
symbols are understood to form a bridge between the experience of the
senses and that which lies or reaches beyond. In this understanding symbols
are referred ultimately to the coexistence of similarity and dissimilarity
between creatures and God, which the Middle Ages conceived as the analogy
of being.1" This was one of the two principal medieval ways to think of the
world-God relationship, the other being participation tending toward unity.
Except in the context of the Pseudo-Dionysian tradition, however, the
Latin term "symbolum" was rarely used in the Middle Ages to refer, not to
the Creed, but to the more general concept of symbol. Apparently other
terms and concepts were considered more appropriate for this meaning in
the Latin West. The most important of these was the term "signum," sign.
For Origen sign and symbol could be synonymous terms. In St. Augustine's
De doctrina christiana the term signum completely absorbed the meaning of
symbol.12 Among the several fundamental distinctions that Augustine made
in this connection, two are most important. The first is the distinction
between signa naturalia and signa data,13 the latter being "given," that is to
say, instituted by man or by God. In the Middle Ages, man-made signa data
are, for example, the so-called insignia, political or ecdesiastical signs of
rulership or office such as royal or imperial crowns, sceptres, mantles, etc.,
episcopal and papal mitres, tiaras, staffs, and liturgical or nonliturgical
nell'alto medioevo, Settimane di Studio del Centro Italiano di Studi sull'Alto Medioevo 23, 1
(Spoleto, 1976), pp. 33 ff.
9 For the earlier use of the Pseudo-Dionysian term symbol by John Scot, see M. Cappuyns,
O.S.B.,Jean Scot Erigeme (Louvain, 1933; repr. Brussels, 1964), p. 295, notes 1 and 2, and E. De
Bruyne, Etudes d'esthetique medievale, 1 (Brugge, 1946), pp. 342 ff.; Pepin, op. cit., passim.
10 Hugh of St. Victor, Commentar. in Hierarchiam Coelestem S. Dionysii Areopagitae 2 (to cap. 1),
PL 175:941B: "symbolum est collatio formarum visibilium ad invisibilium demonstrationem."
11 That in the analogy between God and creatures similarity can never be so perfect as to
"catch up" with dissimilarity, was expressed as follows in the Acts of the Fourth Lateran Council
(1215), cap. 2, Denzinger-Umberg, Enchiridion Symbolorum, 21st-23rd ed. (Freiburg i. B., 1937),
202: ". . . quia inter creatorem et creaturam non potest tanta similitudo notari quin inter eos
maior sit dissimilitudo notanda."
12 See especially De doctrina christiana 1.4 (1.2.2), CSEL 80, p. 9; 2.1-5 (2.1-3.4), ibid., pp. 33
ff.; 2.32 f. (2.10.15), ibid., p. 42; 2.57-65 (2.16.23-25), ibid., pp. 49 ff. (on the various kinds of
signa translata and on symbolical numbers); 3.87-91 (3.29.40-41), ibid., pp. 103 f. (on tropes).
See also J. Chydenius, "La theorie du symbolisme medieval," Poetique 21 (1975), 322-327.
'3De doctrina christiana 2.2, p. 34: "Signorum igitur alia sunt naturalia, alia data. Naturalia
sunt quae sine voluntate atque ullo appetitu significandi praeter se aliquid aliud ex se cognosci
faciunt, sicuti est fumus significans ignem"; 2.3, ibid.: "Data vero signa sunt quae sibi quaeque
viventia invicem dant ... et signa divinitus data quae scripturis sanctis continentur, per homines
nobis indicata sunt, qui ea conscripserunt."