In many places, the Scriptures tell us that all the world is created and upheld according to a wise Creator’s plan. In Psalm 119, for instance, David tells us that all things are God’s servants, existing by his appointment (v. 91). Psalm 104 teaches that God made and ordered all things in wisdom and for a reason. (See also Job 38, 39; Ps. 139:13-14; Eccles. 3:11; Is. 45:18; Jer. 33:20; Col. 1:17; etc.) And the writer to the Hebrews declares that the Son of God continues to “uphold the universe by his word of power” (Heb. 1:3).
The Christian scholars of the Middle Ages dealt with the idea of creation as a matter of course. In the thirteenth century, for instance, St. Bonaventure taught that the purpose of creation is the communication of divine glory, and that the perfection of God is the only satisfactory explanation for the structure of the world.1 Etienne Gilson explains Bonaventure’s teaching by saying, “The science of nature is, as it were, the ethics of things. . . . For every creature, . . . it is one and the same thing to exist and to praise the Lord . . . [for] God created the universe as an author composes a book, in order to manifest His thought.”2 Within the intellectual limitations of our humanity, we are able, by reading this book, to discern the mind of God, because each part of the book bears to some extent the image of God, an image which is both an effect of the Creator’s agency and an analog of some aspect of the Creator’s being.3
Music was by no means exempt from the medieval course of instruction and in fact played a central role in the schools’ and universities’ program of relating everything known to God’s power and wisdom. All medieval students in the Christian west started in higher education by learning the three arts of the trivium: grammar, logic, and rhetoric. They proceeded then to the four numerical arts of the quadrivium: mathematics, geometry, astronomy, and music. In this curriculum, music theory itself was considered a Christian study, appealing to other Christian truths as analogies or premises, and in turn was considered one of only seven foundational studies prerequisite to doctoral training in theology, medicine, philosophy, or law.4
More recently, in The Idea of a University, John Henry Newman eloquently defends this longstanding Christian tradition, explaining that all parts of a proper knowledge of our world ought to be integrated. “All branches of knowledge,” Newman says, “are connected together, because the subject-matter of knowledge is intimately united in itself, as being the acts and the work of the Creator. Hence it is that the Sciences, into which our knowledge may be said to be cast, have multiplied bearings one on another, and an internal sympathy, and admit, or rather demand, comparison and adjustment. They complete, correct, balance each other.”5 Newman encourages scholars to develop “enlargement of mind,” which he describes as “the power of viewing many things at once as one whole, of referring them severally to their true place in the universal system, of understanding their respective values, and determining their mutual dependence.”6
As we know, the academic culture is almost completely different today. Many Christians (indeed people of many other faiths, as well) claim to encounter God in musical experience. And yet the appeal to God as an explanation for any musical fact or phenomenon is not to be found in current academic journals. The situation creates the appearance that all those claims of hearing God in music are intellectually indefensible. It is not my purpose to rehearse changes in intellectual culture during the modern era or the reasons for those changes but rather to explore the possibility of semiotics as a foundation for a methodologically rigorous defense of the claims of hearing God in music, a foundation that could make such claims acceptable in today’s academic climate.
Semiotics is generally defined as the study of signs. When semioticians speak of “signs,” though, they do not just mean road signs and messages on post-it notes. Most semioticians tell us that everything is, for us, a sign: everything means something else. They may disagree on the nature of reality, that is, on the definition of “thing,” but what ever the thing is-concrete object, idea, fragmented totality of intersubjective experience, or solipsistic dream-the thing is a sign. We could say, then, that semiotics represents an attempt by modern (and postmodern) humanity to grapple with the interconnectedness of the world. In the words of musical semiotician Jean-Jacques Nattiez, “An object of any kind takes on meaning for an individual apprehending that object as soon as that individual places the object in relation to areas of his lived experience.”7
The similarity between Nattiez’s words and Newman’s is striking. Both encourage us to understand everything we encounter in relationship to everything else we know. The ideal Christian following Nattiez’s program would place each new experience in relation to everything else, contiuously appreciating more and more the role each part plays in what Newman calls the “universal system,” ultimately seeing the complex whole as testimony to the power, wisdom, and goodness of the Creator. The semiotic view that all phenomena are received as signs is also fully compatible with the scriptural teaching that Nature speaks. (Ps. 19:1-4; Ps. 42:7; Ps. 148; etc.) We might remember here Bonaventure’s teaching that “for every creature, . . . it is one and the same thing to exist and to praise the Lord.”
It may come as no surprise, however, that the field of semiotics offers some ideas that Christians will not be able to embrace. For instance, while Bonaventure speaks of creation as the communication of God’s glory, Nattiez tells us that signs do not communicate, that they do not transmit content from an author to an audience, and that semiotics only studies the production, structure, and reception of signs, and the continual reconstruction of meaning.8
Going further, some semioticians deny the existence of meaning. Bonaventure tells us God is the “transcendent finality,” in Gilson’s words again, “the law which defines the creature’s structure.” Similarly, semioticians speak of the hypothetical “transcendent signified,” absolute, stable, and timeless. In author Daniel Chandler’s words, “All other signifieds within that signifying system are subordinate to the dominant central signified which is the final meaning to which they point.”9 Jacques Derrida, however, argues that the transcendent signified is only a tool of dominant ideology, and is he not alone in denying the actual pertinence or accessibility of the transcendental.10
Going further yet, some flatly deny the existence of God. Recall that Bonaventure characterized God as the Author of the book known as Creation. In his famous essay, “Death of the Author,” Roland Barthes makes the same connection between God and author, world and text, but argues that no text (including the world) has inherent meaning, that no author and no God exist. “Literature, by refusing to assign a ‘secret’, an ultimate meaning, to the text (and to the world as text), liberates what may be called an anti-theological activity, an activity that is truly revolutionary since to refuse to fix meaning is, in the end, to refuse God and his hypostases-reason, science, law.”11
Most introductory discussions of semiotics (also known as semiology in some European scholarship) will trace the field back to Cours de linguistique générale, an influential book compiled from notes of lectures by the early twentieth-century Swiss linguistics professor Ferdinand de Saussure. Saussure described a sign as a combination of a signifier and a signified. According to Saussure, both signifier and signified are ideas, mental phenomena: a sign does not connect the word “tree” and a real-world object with branches and leaves, but rather the sound-image of the word “tree” and our mental concept of “tree.” Important to Saussure’s system is the notion that this relationship is totally arbitrary. For us the sound “tree” calls to mind the concept, but experience could condition us just as easily to associate the sound “arbre” with the concept, or the sound “sproink,” for that matter. If the relationship between signifier and signified is conventional, the sign’s real significance must lie elsewhere: Saussure teaches that a sign’s significance is determined by its systemic relationship to signs that come before and after it and to other signs that could replace it. What words could precede “tree”? “The” perhaps, or “tall,” but not “want.” In the sentence “I see a tree,” what words could replace “tree”? “Dog” would work, but not “red.”
Author John K. Sheriff offers an important critique of the ramifications of Saussure’s view, ramifications that come to light in a history he outlines as follows. Adopting the Saussurean view of the sign, structuralists such as anthropologist Claude Lévi-Strauss posit the claim that the meaning of anything we do or say comes not from someone speaking, writing, acting, composing, or performing-not from a person but from a system: from cultural conventions or from language and from the structures these systems make possible. Taking meaning as a given, structuralist theorists try to codify the systems that make meaning possible, and structuralist critics such as literary critic Monroe Beardsley seek to determine the meaning of individual works according to the structure of the given text alone, without reference to the author.
But, Sheriff notes, critics from later in the twentieth century such as Roland Barthes and Umberto Eco point to the importance of the individuality of the reader in the interpretation of a text. Not only is meaning independent of the intentions of a speaker, writer, composer, and so on; it is also not entirely fixed by the structure of a given text. Instead, says Barthes, every reader interacts with a text using a unique set of interpretive systems. Paul de Man goes a step further by pointing out that each reader is constantly changing, even as he reads a given text. Sheriff summarizes this bleak history by noting that first the author becomes unapproachable, then the text, and finally the experience of the text.
Because of our structuralist assumptions, whatever we wish to study eludes us. Everything that may have seemed objective, given, or having significance in itself is really already a part of a system that we brought to the perception of the thing perceived. What we come up with is exactly what our assumptions allow us to come up with-a sign within a system-and the significance of the sign changes as the system changes.12
Jacques Derrida, then, offers deconstruction, the view that, in Sheriff’s words, “all meaning is supplementarity, an ideality exterior to the process of language.”13 In an infinite chain of deferral, signs only represent other signs; they never point to reality. Madan Sarup characterizes this post-structuralist view as teaching that, “there is a perpetual detour on the way to a truth that has lost any status or finality.”14 As a result, we are left with a choice: we can either hopelessly seek truth or joyously indulge in the play of interpretation, affirming, in Derrida’s words, “a world of signs without fault, without truth, and without origin which is offered to an active interpretation.”15
At the end of this historical review, Sheriff points out that the road to radical uncertainty began with the step of accepting Saussure’s model of the arbitrary sign. An alternate view is offered by nineteenth-century American philosopher Charles Peirce. In Peirce’s semiology, a sign is a representamen, a signifier that stands to somebody for some thing, an object. The representamen acts by invoking in the mind a new, more developed sign called an interpretant. Now while Peirce’s interpretant is always a mental image, the representamen and the object are not necessarily mental images, as is the signified in Saussure’s scheme; the assumption is made from the outset that a sign can be a part of external reality and that it can represent something in an external reality. In addition, Peirce, anticipating and arguing against Derrida, understood a pragmatic need for an end to perpetual deferral: all signs pointing only to other signs. W. B. Gallie characterizes Peirce’s thinking in these words: “As Peirce frequently points out, the exigencies of practical life inevitably cut short such potentially endless development. Moreover . . . , any merely random succession of a sign’s potential interpretants . . . would evidently lack what we ordinarily refer to as direction, point, and purpose.”16 Because I believe in absolute reality, share the assumption that signs can represent that reality, and agree with the propriety and the need to avoid the infinite regression of interpretation, I adopt the Peircean semiotic model.
Peirce describes three modes of signs: depending on how the sign relates to its object, it is either an icon, an index, or a symbol.17 In Peirce’s words, the icon “exhibits a similarity or analogy to the object” and “partakes in the character of the object.” Examples include a stick figure to represent a person, and the spoken word “quack” to represent the sound of a duck. The index, says Peirce, “forces the attention to the particular object intended without describing it” by virtue of some natural relationship to what it signifies such as a physical connection or causation. Examples include smoke as a sign of the presence of fire, and the sound of quacking as a signal that ducks are nearby. The symbol, by comparison, cultural and arbitrary, “signifies its object by means of an association of ideas or habitual connection.” Examples include clapping as a sign of approval and almost all words, including the word “quack” as a designator for a fraudulent or incompetent physician. This quick summary points up another reason for a Christian to prefer Peirce’s model over Saussure’s: the recognition that the connection between a sign and its object is not always arbitrary. In Bonaventure’s doctrine, the image of God born by each element of creation relates to God either as an analogy or as an effect; in Peircean terms, they point to God in either the iconic mode or the indexical mode. Signs originating with the all-wise God can never be arbitrary.
Having covered some basic concepts and terminology, we are now ready to apply the ideas to music. If we are to use semiotics to defend the notion of hearing God in music, we must define the way in which music acts as a system of signs. Indeed, several authors have proposed semiotic systems for the analysis of music. Sadly, though, all seem inadequate for the task at hand: providing a theoretical foundation for saying that a musical piece or experience makes reference to God. The most promising of the group is Jean-Jacques Nattiez. I’ve already expressed my guarded concerns about Nattiez’s denial of communication and the reconstruction of meaning. But this is not to deny the value in Nattiez’s view, which places several parts of the musical process in their relative places. In his book Music and Discourse, Nattiez describes what he calls a tripartite view of musical discourse: music exists at three levels. Through a poietic process (the first level), a composer produces an artifact, a piece of music. This artifact, then, has a structure of its own which exists at the second level: what he calls the neutral level. At the third, esthesic level, a receiver-a listener, that is, or a performer or analyst-performs receptive acts on the artifact. Among the aspects Nattiez does not address in this tri-level view are the ultimate identity of the meaning of a piece and the roles of nature and God in the process.
Philip Tagg, in almost diametrical opposition to Nattiez, stresses communicated intentions and audience reception at the expense of thoroughness in the analysis of the artifact.18 More importantly, his system recognizes only (human) culture and artifact, not God.
Kofi Agawu’s book Playing with Signs, offers a semiotic approach that combines Leonard Ratner’s topical analysis with a formalism influenced by the theories of Heinrich Schenker.19 He offers interesting, rich analyses; but for our purposes, the sources of the topical references are all too human, too conventional, while the significance of structural lines and syntactical paradigms is simply accepted, their possibly transcendent origins unexplored.
Eero Tarasti’s book A Theory of Musical Semiotics presents a methodology for analyzing what he refers to as modalities in music: being, doing, wanting, knowing how to, believing, etc.20 Tarasti, however, does not show interest in identifying either objects or the events to which these modalities relate. What does a desiring phrase want? What does a knowledgeable phrase know how to do? He never says. The empty formality of signification calls to mind Derrida’s infinite deferral of meaning and endless play of signification without truth; thus Tarasti offers little to the present endeavor.
Robert Hatten’s work describes semiotics as a happy mean between mathematical approaches to music and metaphysical approaches.21 The mathematical approaches, he says, deal with natural objects and human mental states. But what things constitute the proper object of the metaphysical analysis? Hatten does not say. By omission, he implies that such approaches are objectless and that a world transcending natural objects and human minds does not exist. The Christian, to recognize the proper place of Creation, needs a metaphysical grounding that acknowledges natural objects and human mental states to be the product of a divine mental state.
While each of these approaches taken as a whole fails to serve our present needs, we certainly may draw helpful elements from them. Placed on a Peircean foundation, these ideas can indeed help explain how it is that people find music signifying God. I recommend a discerning use of semiotic theories and propose a model, drawn primarily from the taxonomies of Peirce and Nattiez, for semiological analysis of music as a way of viewing any given musical moment (a piece, a style, a structure) according to its “true place in the universal system.” As it happens, Peirce’s three modes of signification – index, icon, and symbol – line up rather neatly with Nattiez’s three levels of musical discourse: poietic, neutral, and esthetic.
First, in analysis of what Nattiez calls the poietic level, the Christian is able to consider intentions in a way that respects the personhood of the composer (whereas structuralist thinking tends to declare authors and composers irrelevant, their intentions irretrievable) and to consider cultural influence in a way that acknowledges tension in the value of culture (whereas Derrida and others tend to see culture in terms of dominant ideologies that bully their way into language). It is at this poietic level that Peirce’s symbolic mode predominates, as man, the image of God, creates works that refer to the extramusical world in culturally determined ways. Recognizing, for instance, the pertinence of the intentions of Handel and, more importantly, his librettist makes for a richer and more ennobling experience of Messiah. When harmonies enter only on the last word of the line “The people that walked in darkness have seen a great light,” it is because a person named Handel wanted to say something about the light. The connection between harmony and light is not natural or based on analogy; it is purely symbolic, arbitrary, cultural. But knowing that a person drew the connection intentionally, the connection works for us. Similarly, where Old Testament prophecies pervade an oratorio about Jesus Christ, it is because a person named Charles Jennens wanted to make an argument that the fulfillment of those prophecies in the life of Jesus simultaneously justifies the miraculous nature of the prophecies and proves that Jesus was and is the predicted Messiah. In addition to restoring the musical work to its proper relation with its makers, viewing the composer and librettist as persons places the creative process and the human nature of the makers in proper relation to God: a composer or librettist only creates a work of art because he is created to follow God’s heart in this regard, to be what Tolkien calls a “sub-creator.”22
Second, analysis of the ordered structure of the music itself, what Nattiez calls the neutral level, allows the Christian to celebrate artistic beauty as a crafted rendition of divine order. Music here acts as a Peircean icon, suggesting divine beauty by analogy while partaking of its character. Contrapuntal structure in a Beethoven symphony, for instance, harmonic progressions in a Bach chorale, and textural delineations in a Beatles song all testify to the ontological primacy of order over disorder in the created world, in turn testifying to the wisdom of the Creator. In Tolkien’s words, “If [the sub-creator] indeed achieves a quality that can fairly be described by the dictionary definition: ‘inner consistency of reality,’ it is difficult to conceive how this can be, if the work does not in some way partake of reality. The peculiar quality of the ‘joy’ in successful Fantasy [or musical composition!] can thus be explained as a sudden glimpse of the underlying reality or truth.”23 In addition, while virtually all contemporary analysis of the piece itself concentrates on this structure, the Christian should not ignore the material arranged by the structure: the natural acoustic material that God gives the composer to work with. When a final harmony rings sweetly in the rafters of a vaulted cathedral ceiling, it works because a wise Creator made it to work. We indeed hear God in this situation because we hear the effects of his creative labor.
The last example leads us to the third pairing of terms from these threefold systems. At Nattiez’s esthesic level, the view of music as perceived experience, the Christian may explore the interplay of cognition, acoustics, and physiology, all parts of the creation that point to God by what Peirce would call an indexical mode of signification: any consideration of natural phenomena raises questions of causality, and the Christian recognizes God as the ultimate Cause. Whereas the first note of Mahler’s first symphony, for instance, can simply be declared a structural dominant, recognition of its ambiguity as first perceived by a listening agent places the moment in a context that more fully recognizes the integral nature of creation. The note is not just a note; it is a note heard by a person who does not know its ultimate significance and yet awaits attentively to discover the truth. We do not know what to make of the note at first because we are finite beings, made to live in time. But we want to know, because it is the glory of kings to search a thing out (Prov. 25:2). When we think this way, we hear God because our thoughts move not in a forbidding world of arbitrary, meaningless mental phenomena but a rich, palpable world in which every musical sound lives at the nexus of God’s laws of physics and receptive minds made in the image of their Creator.
The field of semiotics presents the Christian scholar with a mixed bag of promises and pitfalls. On the one hand, it addresses important issues: language, meaning, society, the nature of the self, communication, and art. On the other hand some of its assumptions have led certain scholars to the denial of truth, the denial of the self, and the denial of God, all ends which the Christian must clearly guard against.
Jacques Lacan speaks of the dominance of the signifier, in his words “the incessant sliding of the signified under the signifier,” with the result that all signs signify nothing.24 The wise bard spoke of a tale told by an idiot signifying nothing, but this tale is told by post-structuralists with academic posts and references to their works in thousands of intellectual journals. Scripture can certainly accommodate the tale if told to explain our intellectual and linguistic limitations. We are told after all that God confused our language at Babel. (Ps. 139:6; Eccles. 3:11; Is. 45:15; Gen. 11:9; etc.) But when we are told that reality itself, including any real God, is shaped by our language, the Christian must show extreme caution.
Yet Christians should embrace truth no matter what the source. In Augustine’s words, “Let every good and true Christian understand that wherever truth may be found, it belongs to his Master.”25 We are called to take captive every thought that is raised against Christ, but when confronted by a body of ideas that is only partially true, instead of cowering in fear or railing against blasphemy, let us follow the example of Paul on Mars Hill, commending the benighted for the glimmer of light they do enjoy, and then simply declaring the name of the unknown God.
Agawu, V. Kofi. Playing with Signs: A semiotic interpretation of classic music. Princeton: Princeton University Press, 1991.
Augustine. On Christian Doctrine. Translated by J. F. Shaw. In Great Books of the Western World, ed. Robert Maynard Hutchins, vol. 18, 621-98. Chicago: Encyclopedia Britannica, 1952.
Barthes, Roland. “The Death of the Author.” In Image, Music, Text. Translated by Stephen Heath. New York: Hill and Wang, 1977).
Carpenter, Nan Cooke. Music in the Medieval and Renaissance Universities. Norman, OK: University of Oklahoma Press, 1958; reprint, New York: Da Capo Press, 1972.
Chandler, Daniel. Semiotics: The Basics. London and New York: Routledge, 2002.
Derrida, Jacques. A Derrida Reader: Between the Blinds. Edited by Peggy Kamuf. New York: Columbia University Press, 1991.
——–. Writing and Difference. Translated by Alan Bass. London: Routledge, 1978.
Gallie, W. B. Peirce and Pragmatism. New York: Dover, 1966; reprint, Westport, CT: Greenwood Press, 1975.
Gilson, Etienne. The Philosophy of St. Bonaventure. N.p.: Franciscan Press, 1965.
Hatten, Robert. “Toward a Semiotic Model of Style in Music: Epistemological and Methodological Bases.” Ph.D. diss., Indiana University, 1982.
Lacan, Jacques. Écrits: A Selection. Translated by Alan Sheridan. New York: Norton, 1977.
Nattiez, Jean-Jacques. Music and Discourse: Toward a Semiology of Music. Translated by Carolyn Abbate. Princeton, NJ: Princeton University Press, 1990.
——–. “Varese’s [sic] ‘Density 21.5’: A Study in Semiological Analysis.” Translated by Anna Barry. Music Analysis 1/3 (October 1982): 243-340.
Newman, John Henry Newman. The Idea of a University. New Haven, CT: Yale University Press, 1996.
Peirce, Charles. Peirce on Signs. Edited by James Hoopes. Chapel Hill, NC: University of North Carolina Press, 1991.
Popper, Karl and John C. Eccles. The Self and Its Brain. New York: Springer, 1977
Sarup, Madan. An Introductory Guide to Post-Structuralism and Postmodernism. Athens, GA: The University of Georgia Press, 1989.
Saussure, Ferdinand de. Course in General Linguistics. 1916. Translated by Wade Baskin. New York: McGraw-Hill, 1966.
Sheriff, John K. The Fate of Meaning: Charles Peirce, Structuralism, and Literature. Princeton: Princeton University Press, 1989.
Tagg, Philip. “Introductory Notes to the Semiotics of Music.” Version 3, July 1999. www.tagg.org/xpdfs/semiotug.pdf.
Tarasti, Eero. A Theory of Musical Semiotics. Bloomington: Indiana University Press, 1994.
Tolkien, J. R. R. “On Fairy Stories.” In The Tolkien Reader, 33-99. New York: Ballantine, 1966.
Van Duesen, Nancy. Theology and Music at the Early University: The Case of Robert Grosseteste and Anonymous IV. Leiden: E. J. Brill, 1995.
1. Etienne Gilson, The Philosophy of St. Bonaventure (n.p.: Fransiscan Press, 1965), 181.
4. Nan Cooke Carpenter, Music in the Medieval and Renaissance Universities (Norman, OK: University of Oklahoma Press, 1958; reprint, New York: Da Capo Press, 1972), 11-13; Nancy van Deusen, Theology and Music at the Early University: The Case of Robert Grosseteste and Anonymous IV (Leiden: E. J. Brill, 1995), ix-xv.
5. John Henry Newman, The Idea of a University (New Haven, CT: Yale University Press, 1996), 76.
7. Jean-Jacques Nattiez, Music and Discourse: Toward a Semiology of Music, trans. Carolyn Abbate (Princeton, NJ: Princeton University Press, 1990), 9.
9. Daniel Chandler, Semiotics: The Basics (London and New York: Routledge, 2002), 245.
10. Jacques Derrida, A Derrida Reader: Between the Blinds, ed. Peggy Kamuf (New York: Columbia University Press, 1991), 34-41.
11. Roland Barthes, “The Death of the Author,” in Image, Music, Text, trans. Stephen Heath (New York: Hill and Wang, 1977), 147.
12. John K. Sheriff, The Fate of Meaning: Charles Peirce, Structuralism, and Literature (Princeton: Princeton University Press, 1989, 26.
13. Ibid., 53.
14. Madan Sarup, An Introductory Guide to Post-Structuralism and Postmodernism (Athens, GA: The University of Georgia Press, 1989), 3.
15. Jacques Derrida, Writing and Difference, trans. Alan Bass (London: Routledge, 1978), 292.
16. W. B. Gallie, Peirce and Pragmatism (New York: Dover, 1966; reprint, Westport, CT: Greenwood Press, 1975), 126-27.
17. See Charles Peirce, Peirce on Signs, ed. James Hoopes (Chapel Hill: University of North Carolina Press, 1991). All quotations are taken from pp. 181 and 251.
18. Philip Tagg, “Introductory Notes to the Semiotics of Music,” Version 3, July 1999, www.tagg.org/xpdfs/semiotug.pdf.
19. V. Kofi Agawu, Playing with Signs: A Semiotic Interpretation of Classic Music (Princeton: Princeton University Press, 1991).
20. Eero Tarasti, A Theory of Musical Semiotics, (Bloomington: Indiana University Press, 1994).
22. J. R. R. Tolkien, “On Fairy Stories,” in The Tolkien Reader (New York: Ballantine, 1966), 49.
24. Jacques Lacan, Écrits: A Selection, trans. Alan Sheridan (New York: Norton, 1977), 154.
25. Augustine, On Christian Doctrine, trans. J. F. Shaw, in Great Books of the Western World, ed. Robert Maynard Hutchins, vol. 18 (Chicago: Encyclopedia Britannica, 1952), 646.