Lambeth Fragments: Language of Being and Nonbeing
Linguistic Analysis, Method, and Pedagogy Behind this diverse set of tactics, Anselm's linguistic observations are governed by a principle derived neither from logic or grammar: that the orders of things and words are not identical. Language, whether grammatical or dialectical, can mislead us. There are, in other words, many ways of going wrong with language and perhaps (to further parody
Aristotle on virtue) not even one way of going right. Anselm thinks there are all kinds of problems with language: we are imprecise in our use and language itself is ambiguous and misleading.83 It is this aspect of Anselm on which Henry focuses that supports the connections among Anselm's twentieth-century linguistic analysts like Frege, Tarski, Carnap,
Russell, Wittgenstein, Reichenbach, and Peirce.84 But language for Anselm is also, as Bencivenga writes, "incomplete;" "it does not cover reality; it leaves things large and small totally unaccounted for, speechless; it forces us to stretch our resources, to use whatever comes close, inappropriate as it may be."85 Those focused on Anselm's linguistic analysis do not take this last step and, therefore, do not connect Anselm's concern with language adequately with his metaphysical and theological concerns. Anselm's clarifications of what ordinary language means but does not quite say are ultimately in the service of saying what we mean but cannot quite say (or understand) about God and his creation. And, as we shall see throughout Anselm's writing, Anselm does not think that closing the gap between what we mean and what we say is possible in every respect—whether the topic is God or creatures. Thus, these commentators miss the ways in which in Anselm's examination of finite and infinite reality reaches toward necessary reasons and linguistic breakdown.
Anselm's aspirations, for which language is the means, in other words, cannot be satisfied by logical or grammatical analysis.
Lambeth Fragments: Language of Being and Nonbeing The unfinished work known as the Lambeth Fragments is concerned about the same thing the De grammatico is but in a more general and disturbing way.86 In these fragments, Anselm gives a complex and sophisti83. Bencivenga, Logic and Other Nonsense, 18–19.
84. Bencivenga, Logic and Other Nonsense, 21–24, 28–30.
85. Bencivenga, Logic and Other Nonsense, 19. For Bencivenga, the crucial question about Anselm is how he can engage in the project of linguistic analysis and the search for necessary reasons for that which, Anselm readily admits, is ineffable: how, Bencivenga asks, can you strive toward a goal that you do not know? How can you even know you are on your way toward it? (Bencivenga, 30–
32). It is the right question. However, Bencivenga's answer, though extremely creative and thoughtprovoking, is somewhat cryptic and not a thorough look at Anselm's work as a whole and in his time.
86. Henry, "Saint Anselm's De Grammatico," 118–20.
Linguistic Analysis, Method, and Pedagogy cated catalogue of the ways in which we do not say exactly what we mean and of the ways in which language in the most ordinary and common expressions is never, or hardly ever, proper. The focus is not on nouns, as in De grammatico, but on modal expressions and verbs. In these short pieces, Anselm tries to systematize degrees of impropriety and links complexities regarding verbs to the issue of improper language in general.87
This text both exemplifies Anselm's philosophical method of "linguistic analysis" and forms the connection between the apparently purely logical concerns of De grammatico and the rest of Anselm's theological and philosophical work. For in these fragments the method is linguistic, but the work suggests connections between the analysis of language and notions of freedom, possibility, necessity, causality, and power central to the problems Anselm takes up in the great theological treatises, Monologion, Proslogion, and Cur Deus homo, and in the discussions of free choice and evil in the dialogues.
For Henry, Anselm highlights improper language in these fragments in order to translate it into proper language, that is, the language of logic, thereby solving—or at least forging the basic tools necessary for solving—the substantive philosophical problems he grapples with in the treatises. But as we shall see, the victory of proper over improper language, of logical over ordinary language in the Fragments, as in De grammatico, is somewhat less than complete. First, for Anselm language used to describe creatures can be used of God only by stretching standards of linguistic propriety almost beyond recognition. Second, Anselm shows that even ordinary language used for ordinary things cannot be purged of ambiguity in favor of a pure and proper technical language. We want and need to use words like "cause," and "can," and, of course, "be" in the different modes he enumerates. These expressions should be used advisedly,
Anselm seems to be saying, an end he aids by listing and distinguishing these derivative modes. Thus, while it is true that the Fragments help link
Anselm's linguistic propensities to his larger philosophical and theological concerns, it does not link them by reducing the problems they explore to problems of language.
Though De grammatico opens with a kind of paradox, that grammaticus is both a substance and a quality, this is a fairly narrow technical prob87. LF 34, 29–40. These fragments have been published in two editions, one edition following the order of the original manuscript (in Memorials of St. Anselm, edited by F. S. Schmitt and R. W. Southern [London: Oxford University Press, 1969], 333–51), another attempt to arrange the fragments in a logical order (edited by F. S. Schmitt in Ein neues unvollendetes Werk des hl. Anselm von Canterbury, in Beiträge zur Geschichte der Philosophie und Theologie des Mittelalters, 33/3 [Münster:
Aschendorff Press, 1936]). Though I have used the latter edition, none of my conclusions depend on this being the actual order intended by Anselm.
Linguistic Analysis, Method, and Pedagogy lem. These unfinished logical, grammatical fragments, by contrast, consider problems and paradoxes worthy of the speculative treatises. Some of the most important of these concern modal notions. For example, the student brings up the paradox of making predications of what does not exist: "that which does not in any way exist has no ability (potestatem).
Hence, it has neither the ability to be or not to be."88 Another fragment considers the problem of denying God's ability to do wrong: we say that lying is impossible to God and justice necessary, but "impossibility carries [with it the sense] of impotence and necessity of violence," both of which would be contradictory to the divine nature.89 "If," the student argues, "you respond that this impossibility and this necessity signify an insuperable strength, then I ask: why is this strength indicated by names signifying infirmity?"90The knowing student who poses these problems is aware there is an inconsistency at the root of these paradoxes, but he cannot find it.91
The dialogue is an unusual combination of arguments and observations. On the one hand, the text considers matters which are highly germane to Anselm's most important and substantive philosophical concerns—the nature of God, the nature of God's actions in the world, and the nature of human responsibility. On the other hand, it engages in listing linguistic complexities and distinctions seemingly for their own sakes. It is not clear whether these fragments were supposed to be part of a single unfinished work or whether there were notes on separate problems destined for different works. Moreover, Anselm never directly solves these paradoxes, yet we can get some ideas about how he might have by looking at the distinctions he makes concerning facere ("to do" or "to cause"), "will," "cause," "something," and "power."
For "cause," Anselm enumerates modes of causing or doing that fall away by degrees from pure action and pure causality:
We say of any given thing that it causes something to be either because it causes the very thing which it is said to cause to be, or because it does not cause this very thing not to be, or because it causes something else to be, or because it does not cause something else to be, or because it causes something else not to be, or because it does not cause something else not to be.92
"Cause," then, can be said in ways that cover pure and direct action, pure and direct inaction, as well as indirect action and indirect inaction.
Anselm also supplies four modes in which "something" can be said.
88. LF 23, 13–14.
89. LF 24, 16–19.
90. LF 24, 22–25.
91. LF 24, 14–15.
92. LF 29, 27–30.
Linguistic Analysis, Method, and Pedagogy The first and proper mode is when the thing has a name, is conceived by the mind, and exists in reality (e.g., animal). In the second mode, a thing has a name, is conceived by the mind, but does not exist (e.g., chimera).
In the third mode, a thing has a name, but no concept and no existence (e.g., injustice and nothing). Lastly there is that which has no name, concept, or existence (e.g., nonbeing, as in "the sun's not being above the earth causes it not to be day").93 The scale is from correspondence between being, thought, and language toward their increasing disjunction.
The verb "to be" also has many modes, and many of these modes, Anselm notes, can be found in other verbs as well.94 The modes for the verb
"to be" are modes of improper speech. We say improperly that A is B, Anselm explains, because A is similar to, is the cause, effect, genus, species, whole or part of, has the same capability, the same shape, is signified by or signifies, is the content or container of an object which is properly called B.95 This list ranges from more or less ordinary predications (and ordinarily considered to be proper predications) of the genus or species of an individual, to the metaphorical, that is, predications based on similarity or signification.
In the case of facere (to do), the trend in these analyses reaches its apogee. As Anselm lists meanings or uses for facere he simply observes that it is used in place of every other verb, even its opposite.96 Anselm supports this conclusion in two different ways. First he notes that any verb can be used in the answer to the question, "what is he doing?" This is true, he points out, not just in the obvious cases, when what someone is doing is singing or composing, but also when the answer is that someone is being, living, being able (posse), calling (vocare), or owing (debere). Thus we can answer the question, "what is he doing?" by saying "he is in church,"
"he is living as a good man lives," "he is powerful (potest) over the domain in which he lives," "he owes (debet) much money," and so forth.97 At first it seems that facere, like its modern French equivalent, faire, seems to have no meaning common to all the expressions in which it is used. The Wittgenstinian solution, of course, is that there is no "meaning," no essence uniting the uses; one simply has to learn the appropriate circumstances for all the different expressions. On this view, there is no meaning apart from use, not just in the case of a few common verbs like "do" but in everything we say. Anselm does begin reflecting on this apparent linguistic oddity by noting that "every cause in common usage is said 'to do' that of which it is a cause."98 He offers a general principle to explain this situation: "everything
93. LF 42, 22–43, 23.
94. LF 34, 16–23; 34, 40–41.
95. LF 34, 29–39.
96. LF 25, 13–19.
97. LF 25, 23–26, 4.
98. LF 26, 6–7.
Linguistic Analysis, Method, and Pedagogy of which a verb is predicated makes (facit) what is signified by the verb."99
The odd cases, like "being" and "living" as kinds of doing, help prove the general point, Anselm explains. For we see that the subject of which the verb is predicated is a cause in those cases especially, in the sense that without their agency, whatever activity is being named by the verb would not take place. For example, without the one who endures or suffers (the opposite of doing or causing) there would be no enduring or suffering.100
In another fragment Anselm offers another way of explaining this oddity of ordinary language (that "to do" can stand for all verbs). "Man" is a cause of the things of which it is predicated being animals. "Of whatever
'to be' is said, either simply, as in 'man is,' or with addition, as in 'man is an animal' or 'man is healthy,' the conception of this thing precedes and is a cause of it being said to be or not to be and is [a cause of] what is said being understood."101 Anselm is reflecting on the nature of verbs more generally, locating their function as a kind of doing or causing of that of which they are predicated. This is exactly the kind of explanation Wittgenstein seems to reject in the Investigations. For Anselm, ordinary language makes sense and not just to those dwelling within it as a form of life. To discover that sense, he attempts to analyze "do" and "cause" and "be" into different but ordered modes. Thus he makes a gesture towards understanding how the "common usage" of facere might be understood in a more general way to have some sense behind it. Moreover, this, he implies, will be a way into understanding the grammatical category of verbs, and, further, be woven into one of the metaphysical themes most explored in these fragments, the notion of causality. Language will stand up to philosophical analysis, Anselm believes, but it takes a great deal of careful reflection on how we use language and what we mean in order to see that sense.
But just as Anselm does not enshrine ordinary language and eschew further analysis, neither does he, like some early analytic philosophers, want to rewrite it and free it from all ambiguity. Rather Anselm wants to analyze and order the ambiguities of ordinary language while retaining it and its nuances. Thus Anselm attempts to make what we normally say intelligible rather than nonsense by arranging it on a scale from the most proper to most derivative. If we expect pure univocation, ordinary language (not to mention theological language) will look like a complete failure. But if we have a kind of ordered equivocation in which the less proper and the improper uses are analyzed and ordered, then we both save the language we already use and have one which can tolerate the rigors of theological subjects without completely falling apart.
99. LF 26, 7–9.
100. LF 26, 12–15.
101. LF 28, 5–9.
Linguistic Analysis, Method, and Pedagogy What is intriguing in these fragments is how grammatical and logical concerns are interwoven with philosophical and theological ones. In some cases, observations about language are merely juxtaposed to theological problems or scriptural citations. In others, the analysis seems to be directly in the service of major theological problems raised by scriptural passages. In still others, there seems an equal interest in analyzing language for its own sake and using language to explore a philosophical or theological problem. A good example of the first type is the discussion of the verb facere. For after the observations outlined above, Anselm justifies this broad use of the term by an appeal to scripture: "the Lord in the Gospel uses 'facere,' or 'agere,' which are the same, for all verbs, when he says 'All who do evil hate the light,' and 'he who does the truth comes to the light.'"102 He goes on to explain that "doing evil" encompasses a great deal which does not seem like action, like being present or absent where one ought or ought not be, for example.
There is another example of this kind in the text, concerning verbs which have other verbs as their objects. As an example, Anselm cites a passage from Exodus where a verb appears to function as an object but does not really: "The people sat down to eat and to drink, and they rose up to play."103 Here there is no substantive discussion of the passage or any questions it might raise. The juxtaposition of ordinary language and the language of scripture in this case seems to be the result of such great but casual fluency in the language of scripture that the passages simply suggested themselves to Anselm as any other example from ordinary language would. By contrast, the passages from John about "doing evil" and "doing the truth" are not incidental examples but are the focus of Anselm's discussion of truth and moral action in De veritate.104 De veritate along with the other two dialogues in the trilogy, De libertate arbitrii and De casu diaboli, as we shall see, are profound meditations on the meaning of these and other scripture passages in which Anselm makes use of but expands on the grammatical observations educed in the Fragments. But even in the fragments, Anselm's reflections on the ambiguities of "doing" function as a warning to dialecticians and those in charge of spiritual direction or moral analysis. One should not be misled by the ambiguities of language into circumscribing the range of responsibility to "action" in the grammatical or even natural sense but realize there can be moral responsibility for and even an agency in some kinds of passivity.
Another fragment discusses causality in order to show that there are 102. John 3:20–21; LF 28, 13–16.
103. Exodus 32:16, LF 34, 2–3.
104. DV 5, S I, 181.
Linguistic Analysis, Method, and Pedagogy different kinds of causes that can be simultaneously responsible for the same effect. Anselm carefully develops two sets of divisions of cause into types. He names efficient and material, but also six modes of causality: 1) causing to be the very thing which it is said to cause, 2) not causing this very thing not to be, 3) causing something else (other than what it is said to cause) to be, 4) not causing something else to be, 5) causing something else not to be, 6) not causing something else not to be.105 Anselm notes that this classification can be expanded from applying these categories to nonefficient as well as efficient causes and to all the other verbs joined to facere besides esse.106 Anselm points out that a guard and Herod are both equally said to cause the death of John the Baptist. The one who performs the act kills, but so does the one who orders the killing; the reason for which someone is killed is also a cause, and so on. Agents cause things both by acting and not acting; even something nonexistent can function as a cause.107 Human beings are causes (and, hence, responsible) even when they are not the only causes. There is a converse corollary.
While finite beings always act and are acted on in conjunction with other causes, remote and proximate, God has no cause and at least the first effect of his causality had no other causes.108
Two other questions discussed thoroughly in the Fragments are the meaning of "ought" and the different modes of willing. Anselm distinguishes four modes of willing. The first and most proper sense is the efficient will, which "insofar as it can brings about what it wills."109 The approving will is for what one wishes to happen but does not cause; the conceding will is for what one does not wish but does not object to; lastly, the permissive will is that which neither causes, approves, nor concedes but allows to happen what it could prevent.110 These distinctions are then applied to scripture. When scripture says, "He has done whatsoever He has willed to do," it is describing God's efficient will; when it says, "He hardens whom He wills to," it is describing God's permissive will; and when we read that "God wills for every man to be saved," this refers to God's approving will.111 The way Anselm makes these distinctions give the impression that they were designed to deal with passages in scripture and problems attendant to them. The passages do not simply function as illustrations of the distinctions in the way the citations of scripture passages supporting the account of facere do. But in both types of passages, Anselm exhibits a
105. LF 29, 20–30.
106. LF 32, 21–30; 33, 30–33.
107. LF 40, 8–12; 41, 33.
108. LF 41, 3–16.
109. LF 38, 26–7.
110. LF 38, 26–39, 5.
111. LF 39, 6–26. Anselm cites Ps. 113:3, Rom. 9:18, and 1 Tim. 2:4.
Linguistic Analysis, Method, and Pedagogy kind of fluidity, moving between ordinary and scriptural language and illustrates how he connects linguistic and theological problems.
The topics mentioned in the Fragments, the nature of power, necessity, and willing in God, and the nature of agency, causality, and responsibility in both God and human beings are, of course, discussed in Anselm's other works.112 One might argue that the linguistic analysis developed in the Fragments was designed to be used to resolve contradiction and dissolve apparent incoherence on these issues. Anselm does use them in this way, as we shall see. But there is another, perhaps deeper connection between these fragments and the treatises. The ordering of the many to the one used in the Fragments is a strategy that returns in the Monologion and Proslogion. In the Monologion Anselm argues from lesser to greater levels of goodness to God as the source of goodness. Understanding "that than which none greater can be conceived" requires an ordering of what it is better and worse to be. In the Monologion and Proslogion Anselm confronts the paradox that that which is the source for and deserves the name of a given perfection contradicts the parameters for the ordinary and proper attribution of a term. Conversely, he finds on closer examination that ordinary things do not possess fully the qualities of being or causing that are attributed to them. Anselm touches on both this paradox and its opposite in the Fragments. The discussion of divine power and freedom in the Fragments shows the way in which God sets different standards for action than those appropriate to creatures. The discussion of the ways in which things are said to be or cause something when they are only similar to or the indirect cause of that thing shows the way in which things receive names which they do not full deserve.
These fragments reveal Anselm's recognition that there are difficulties confronting predication of both God and creatures. The problems of predicating ability or necessity of what does not exist (and hence has no abilities), of being and agency of that which is only incomplete being and imperfect agency, shows the way in which language seems to attribute too much to creatures, who exist and have whatever they have only in a relative sense. Another set of observations and distinctions, concerning predications of God, shows the way in which the language of creatures, appropriate to their limited possibilities, imperfect freedom, and incomplete being, results in absurdities when applied to God, who is true and complete being.
112. Hopkins and Richardson's translation of the Fragments includes in the footnotes a concordance of passages to other places in Anselm's writing. See Anselm of Canterbury, edited and translated by Jasper Hopkins and Herbert Richardson (New York: The Edwin Mellen Press, 1976), vol. 2,
226–28.