47. See also De Ver. 10 (S i: 190).
peter king for the success of some statements at signifying the world the way it is; see his account of truth described below.
45. Modern discussions bypass Anselm's (T2) and (T3) by specifying the language as part of the Tarski schema, since we are impatient about fixing meanings.
46. Near the end of De Ver. 2, Anselm makes some misleading remarks in which he suggests that (T2) and (T3) are somehow "natural" and "unchangeable" whereas (T4) and (T5) are "accidental" and depend on the occasion of their use. The only sense in which the former are "natural" is by fixing the language and the conventions for signification, and the only sense in which the latter depend on their use is that they may or may not be true depending on how the world is when the statement is uttered.
47. See also De Ver. 10 (S i: 190).
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila
Anselm on modality My aim is to discuss Anselm of Canterbury's use of basic modal terms (necessity, possibility, impossibility) and his interpretation of the meaning of these and some related notions. I will first sketch modal conceptions in philosophical and theological traditions with which Anselm was familiar, and then take a look at some eleventhcentury modal controversies entered in Peter Damian's On Divine
Omnipotence and also discussed by Anselm. The third section deals with Anselm's views against the general historical background, and the last section is about his attempt to sketch the semantics of modal terms. modalities in anselm's sources The main line of the history of modal theories in ancient and medieval times can be described as follows. There are four originally
Aristotelian ways of understanding the meaning of modal terms in ancient philosophy: the "statistical" or "temporal frequency interpretation" of modality, the conception of possibility as a potency, the conception of diachronic modalities (antecedent necessities and possibilities), and the idea of possibility as noncontradictoriness. I will explain below how these modal paradigms occur in Boethius (c. 480–
523), whose works made them known to early medieval thinkers.
Ancient conceptions did not include the view that the meaning of modal terms should be spelled out by considering simultaneous alternative states of affairs. This new idea was introduced into Western thought in early medieval discussions influenced by Augustine's
(354–430) theological conception of God as acting by choice between possible alternatives. Ancient habits of thinking continued to play
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila an important role in scholasticism, however, and the systematic significance of the new conception was not fully realized before the extensive discussions by John Duns Scotus, William Ockham, John
Buridan, and some other fourteenth-century thinkers. Many scholars have paid attention to the similarities between these late-medieval theories and the contemporary possible-worlds semantics for modal logic.1
One of Boethius's Aristotelian modal paradigms is that of possibility as potency or power (potestas, potentia). Some potencies are never unrealized; they are unchangeably and therefore necessarily actualized. When potencies are not actualized, their ends are said to exist potentially.2 Necessarily actual potencies leave no room for the potencies of their contraries, since they could not be realized.3 Other potencies do not exclude contrary potencies. They are not always and universally actualized, but as potency-types even these must show their genuineness through actualization.4 Boethius thought that natural events typically take place so that active and passive potencies interact and these interactions are determined by chains of causes which force the events to take place, though there are also chance events in nature and voluntary acts are not necessitated by external things.5
The Aristotelian frequency interpretation is another Boethian conception of necessity and possibility. According to its temporal version, to be possible is to be actual at some time and, consequently, what always is, is by necessity, and what never is, is impossible.6
If the state of affairs asserted by a temporally unqualified presenttense-type sentence ("p now") is always actual (or non-actual), the sentence is true (or false) whenever it is uttered and therefore necessarily true (or false). A sentence is possibly true only if what is asserted is not always non-actual.7 Boethius calls the necessity by which a sitting Socrates sits when he sits temporal necessity or conditional necessity. Things which are actual at a certain moment are conditionally necessary because of their immutability at that moment of time, though many of these are changeable and contingent when treated without a temporal qualification.8 The unchangeability of present things was the common background of the ancient axiom of the necessity of the present. It was not merely the logical necessity of the temporally definite disjunction "p or not p."
Boethius argues that since Socrates cannot be sitting and not sitting Cambridge Companions Online © Cambridge University Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
at the same time, he cannot be not sitting at the time when he is sitting.9
In explaining how the necessity of what is actual is compatible with indeterminism Boethius builds on Alexander of Aphrodisias' remarks on diachronic modalities, that is, the possibilities and necessities of events before their occurrences. All contingent events are associated with antecedent prospective alternative possibilities, for example it is possible in the morning that I will be far away in the evening, though this is no longer possible in the evening if I did not travel. An antecedent alternative possibility remains possible until the causes of its opposite will be fixed. After this it ceases to be a possibility. These ideas were associated with chance events and free choices in the criticism of the Stoic doctrine of universal determinism.10 Boethius did not develop a theory of simultaneous or synchronic possibilities which would remain intact even when diachronic possibilities have vanished. On the contrary, he insisted that only what is actual at a certain time is possible at that time with respect to that time.
In Philosophiae Consolatio 5.4.19 Boethius writes: "I do not think that anybody would say that those things which are happening now were not going to happen before they happened." Boethius seems to think here that propositions about future contingents are true, if what is maintained by them will happen. In other places he associates the truth of these propositions with the qualifications "mutably" and "indeterminately." The meaning of these terms is not wholly clear.11 Boethius's discussions influenced the standard scholastic theory according to which future contingent propositions are true or false, since only true propositions can be foreknown and revealed in prophecies, but they are true or false in an indeterminate manner in the sense that the occurrences of contingent events are not yet determined in the chains of causes. To know the truth-values of future contingent propositions does not belong to human epistemic possibilities.12 Boethius seems to think that God as an atemporal being can know contingent things through an immediate vision; all times are present to Him (Phil. Cons. 5.6.25–31). Some medieval authors interpreted the distinction between simple and conditional necessity, which Boethius applied to foreknowledge in this context, as a distinction between two ways of how the notion of necessity can be associated with a consequence: either a consequence is necessary
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila or the consequent is necessary. "What God knows necessarily takes place" can be read in accordance with the simple necessity of consequence as "Necessarily: if God knows that p, then p," which is true, or in accordance with conditional necessity as "If God knows that p, then necessarily p," which is deterministic and false. It is not clear whether this is what Boethius meant.13
While early medieval thinkers could find detailed discussions of philosophical modal theories in Boethius's works, Augustine's doctrine of creation made them aware of other kinds of modal ideas.
According to Augustine, God simultaneously created the first things and the seminal reasons of later things out of nothing. The creation was based on an eternal free act of God's perfectly good will, and took place through His omnipotence. God created time in creating movement in the universe. Time depends on movement, and since
God is unmoving, there is no time before creation. Augustine argues for a sharp distinction between time and mutability on the one hand and timelessness and immutability on the other.14 In Augustine's trinitarian theology, the Son is a perfect image of the Father and, as the Word, the seat of the ideas of finite beings which in a less perfect manner can imitate the highest being. The ideas refer to possible actualization in the domain of mutability.15 So the possibilities have an ontological foundation in God's essence. This was the dominating metaphysical conception until Duns Scotus departed from it by introducing the idea of the primary domain of logical possibility.16
Augustine was familiar with the Neoplatonic thought that the power of being proceeds from the One without leaving any genuine level of being unrealized, but he stressed that the world is not necessary and that lots of possibilities remain unrealized.17 In De Civitate Dei
12.19, Augustine criticizes the ancient doctrines which claimed that the only permissible notion of infinity is that of potential infinity. He argued that an infinite series of numbers actually exists in God's mind, and God could create an infinite number of individuals and know each of them simultaneously. Most of these possibilities remain unrealized.
Augustine regards God's omnipotence as an executive power with respect to His free choice which is conceptually preceded by knowledge about alternative possibilities. God could have done other things, but did not want to.18 Augustine's remarks on various possible histories remained sketchy, but the basic idea is pretty clear
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
and was employed in the early medieval doctrine of God as acting by choice between alternatives. This involved an intuitive idea of modality based on atemporal or synchronic alternatives, which was not employed in ancient philosophy. Let us take a look at some eleventh-century discrepancies between the Catholic doctrine of God's freedom and power and the philosophical modal conceptions which were entered in Peter Damian's On Divine Omnipotence and known also to Anselm. peter damian and the dialecticians Gerbert of Aurillac, Abbo of Fleury, and Garland the Computist, among other tenthand eleventh-century logicians, were acquainted with the Categories and De Interpretatione of Aristotle, the Topica of Cicero, the Isagoge of Porphyry, Boethius's commentaries on these works, and Boethius's own logical monographs. Garland's Dialectica summarizes the main doctrines of the sources in six books.19 Some learned people interested in logic and philosophy asked how Christian doctrine stands in relation to the ancient philosophical heritage and how each should take the other into account. The matter was not of merely theoretical interest, as is attested by the controversy between Berengar of Tours (d. 1088) and Lanfranc (1010–89) about the Eucharist. One of the controversial questions pertained to the application of logic to the Christian mysteries.20 Boethius's theological treatises, the Opuscula Sacra, were discussed in the schools alongside elementary logic, and it seems that this coexistence of theology and dialectic was one of the sources of eleventh-century quarrels.21
Lanfranc, who died as Archbishop of Canterbury, was an Italian, and so was Peter Damian (1007–72), an influential monastic writer and counsellor to many popes. In his book On Divine Omnipotence Damian states that some teachers of his day were applying logic to theology in a way which threatened orthodoxy. According to Damian, logic as a verbal art is not competent to teach anything but how to make statements and to operate with consequences (De Div.
Omnipot. 604a–b, 615a). While generally condemning the misuse of logic in theology, Peter Damian was particularly concerned about the views of dialecticians, which, in his opinion, questioned divine omnipotence.
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila The first group of Peter's examples of philosophical necessities, not acceptable without qualification in theology, comprises traditional examples of invariant causal patterns. According to Peter, such invariances remain in the control of their creator (see, e.g., De
Div. Omnipot. 610d–615b). It has been assumed that Berengar of Tours was one of the dialectical theologians whom Peter Damian is attacking.22 According to Damian, God can change the invariant patterns which are called natural necessities and represent the natural consuetudo (611b, 612b, 614a):
Is it a wonder when he who gave the law and order to nature exercises the power of his will on the same nature so that the necessity of nature does not rebel against him, but acts in submission to the laws, like a servant? The nature of things itself has a nature, namely, the will of God, so that as what is created obeys the laws of nature, so, when ordered, it reverently obeys
God's will, giving up its rights. (De Div. Omnipot. 612c–d) Damian thought in an Augustinian manner that natural causes function uniformly on the condition that God does not want an exception from the common course of nature. (Cf. Augustine, De Civitate Dei
21.8.) Another philosophical notion of necessity discussed by Damian is the necessity associated with dialectical consequences or consequences of discourse (consequentia disserendi):
According to the consequence of discourse, if it is going to rain, then it is entirely necessary that it will rain and thus entirely impossible that it will not rain. Thus, what is said about past things can be concluded no less with respect to present and future things, for as what has been necessarily has been, so what is, as long as it is, necessarily is, and what will be necessarily will be . . . Who would not patently see that if these arguments, having the order of the words they have, are accepted, the divine power is shown to be impotent at all moments of time. (De Div. Omnipot. 603a–b, d)
Damian here describes what Boethius called temporal or conditional necessity, stating that, if it is applied to all moments of time, everything seems to be necessary. Even God cannot do anything but what he in fact does. He also mentions that philosophers have introduced the conception of ad utrumlibet contingency, according to which some things may happen or may as well not happen, in order to speak about things that can be otherwise than they are. Damian thought
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
that since this idea was based on the variable nature of things, it did not really solve the problem contained in the claim that at the time when something is, it necessarily is as long as it is (602d–603a).
Whatever one thinks about dialectical views of necessity or contingency in the temporal order, these do not provide any proper way of discussing God's power, because what is created would then be the criterion for determining what is possible for the creator (cf. 612b).
Damian's brief descriptions of what he considered to be the standard philosophical modalities evidently drew, directly or indirectly, on Aristotle's De Interpretatione 9 and Boethius's commentaries on it.23
The starting point of Peter Damian's treatise is the following statement by Jerome: "I will speak boldly; although God can do all things, he cannot raise up a virgin, after she has fallen" (Jerome, Epistulae
1.150). The idea that what is past is necessary was generally accepted by ancient philosophers. Peter Damian did not want to use the expression that there is something God cannot do. Since divine omnipotence is outside time and there can be no changes in God's possibilities to do things, the historical beginning of the existence of Rome does not have any influence on the fact that God can see to it that Rome never exists (De Div. Omnipot. 619a).
Some authors have thought that in Damian's view God can change the past, and the Law of Contradiction does not impose limits on divine power. This kind of argument seems to be found in the following passage: "Contraries cannot belong to one and the same thing together. This is correctly called an impossibility with respect to the limited power of nature, but it should not be applied to the divine majesty" (De Div. Omnipot. 612a–b).
The context of this text is, however, ambiguous in many ways.
Some scholars think that Damian merely wanted to stress that one should not apply the principles of natural philosophy when speaking about God (see De Div. Omnipot. 597b–c).24 Peter Damian takes it for granted that when the conception of God's power is considered problematic, something is wrong with the manner in which the problem is formulated. If it is asked whether God can do bad or harmful things, it is wrongly assumed that such states of affairs are brought about by a power. Following the Neoplatonic view of Augustine and Boethius, Damian states that to bring about things of this kind is not a sign of ability but of inability (610b–d).25 Furthermore, if it
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila is claimed that when an individual event has taken place, God can no longer prevent it, God's power is discussed in temporal terms.
These are inadequate, since there is no temporal order in eternity, where God makes His eternal decisions in a nontemporal way (De
Div. Omnipot. 619a–620c).
Even though Damian refers to the theological conception of alternative providential programs, he does not elaborate upon this idea while discussing the question of God's having power over the past
(600a, 614d). It seems that, while defending the absolute nature of God's power, Damian was led by the Augustinian conception of God's acting by choice, which implies a theory of alternative possible states of affairs, but he did not manage to create a coherent theory of his own. Damian thought that God's choice is eternal and God has the power to realize the actual choice as well as all possible alternative choices. He did not realize (or chose to deny) that the temporal results of the actual choice exclude a choice with incompatible temporal results and that God's real possibilities to actualize alternative choices are restricted by the actual choice.26 traditional modalities in anselm Anselm of Canterbury was a pupil of Lanfranc at Bec and succeeded him as Archbishop of Canterbury in 1093. He was aware of the debates about the relationship between faith and reason and dealt with questions similar to those in Damian's work on omnipotence, though it is not clear whether he was familiar with it. More than
Damian, Anselm was interested in modal conceptions themselves and tried to codify some aspects of the semantics implied in the Christian use of the notions of necessity and possibility. I shall first deal with Anselm's answers to questions also discussed by Damian and then with Anselm's attempt to develop a semantic theory of modality.
Anselm was much more of a philosophical thinker than Damian, as is seen from his formulation of the question of the relationship between divine and natural possibilities. In Chapter 12 of his dialogue De Casu Diaboli, Anselm asks how it was possible for the world to exist prior to its creation. He states that it was possible because of God's ability to create it, but the world itself did not have the capability to be, and could not be.27 In the late incomplete
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
dialogue called the Lambeth Fragments or Philosophical Fragments, Anselm returns to the same topic. The student presents the following problem: if things can be in virtue of their capability to be and if something which does not exist has no capabilities, and as such neither the capability to be nor the capability not to be, it seems to be both necessary and impossible for something which does not exist to be and, similarly, not to be. This is explicated by the statements that "x has no capability to be" seems to mean "x cannot be" or, equivalently, "it is not possible for x to be" or "it is impossible for x to be" or "it is necessary for x not to be." The statement "x has no capability not to be" is explicated in the same manner: "it is not possible for x not to be" or "it is impossible for x not to be" or "it is necessary for x to be."28 While Anselm often deals with these
Aristotelian definitions, he did not develop any more advanced theory of modal equipollences by analyzing the structures of modal statements, as Abelard did by separating modalities de sensu and de re.29 Anselm thought that the student's paradoxes can be solved through his semantic theory of the meaning of modal words, but this answer is not included in the fragments.
The question of how something which does not exist is possibly something is a problem if actualization is considered as an activation of a passive potency which is embedded in an existing subject, but one that had to be faced, for Christians, and so called for a different modal metaphysics. Since God has created the world out of nothing through His omnipotence, the power of omnipotence is apparently a power which does not presuppose a passive potency in the subject on which it acts. When Anselm states that the world did not have a capability to be before its creation, he did not mean that it would be impossible for it to be. Anselm thought that the world cannot begin to be without an external activator, but divine power does not make impossible things possible. This is explained in the early work Monologion where Anselm deals with the realizable model of the world which is not yet created. Anselm's theory of the ontological basis of the possibility of the world is largely derived from Augustine's trinitarian view of the Son as a Word which is the perfect image of the Father and also includes the ideas of less perfect possible imitations of the highest being. Anselm describes the Word as the perfect divine image of itself; this involves an eternal explication of the highest intelligible being as well as of all things which Cambridge Companions Online © Cambridge University Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila can imitate it in various degrees, since the highest being is the source of all beings and can be imitated by lower beings. Therefore things which can be created have a model or form or similitude or rule in the divine Word (Mon. 9, 31–34). In the Monologion Anselm speaks about the models of actualized things in God's thought, but he believed, like Augustine, that God's thought involves a providential model for the actual history and also models things which are not actualized.30
If this was Anselm's view, why did the possibility of the nonexistent world remain a problem? Anselm apparently thought that in the Fragments the student mistakenly assumed that one could specify the modal status of nonexistent beings on the basis of their lack of any potency. Anselm's point is that a thing before it exists can become actual, if there is an external power which can actualize it and its actuality is not contradictory.31 But even then one could ask how a nonexistent thing which can exist and be understood is something as distinct from a merely impossible thing which cannot be understood.32
Like Peter Damian, Anselm also avoided speaking about the restrictions of divine omnipotence and preferred formulations according to which all necessities and impossibilities are subject to God's will:
As already stated, it is said in an improper sense that God cannot do something or does something by necessity. Indeed, all necessity and impossibility are subject to His will, but His will is not subject to any necessity or impossibility, for nothing is necessary or impossible without His willing it to be so, but it is not true at all that He wills or does not will something because of a necessity or an impossibility.33
Thomas Bradwardine, who often quotes Anselm, paraphrases this text in his influential De Causa Dei (1344). Bradwardine was interested in Anselm's formulation while discussing the question of whether necessary and possible truths are in last analysis freely chosen by God (1.14, 209). When referring to Anselm, Bradwardine is speaking about the de re modalities of existing created beings.
These modalities as inherent potencies or inclinations are created and their existence is dependent on God's will. Bradwardine's view of the foundations of the modalities which precede existence was not voluntarist. In order to explain the difference between created and Cambridge Companions Online © Cambridge University Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
noncreated necessities and possibilities, Bradwardine calls the latter absolute modalities. These are not dependent on God's will:
It is not simply necessary for God to will or not to will this part or that part of such a contradiction, but He can will or not will either of them freely. It is therefore evident that even though any such disjunction is necessary by a simple and absolute necessity which is naturally prior to divine will, so that its necessity or truth is not subject to divine will or to a liberty with respect to contradictories or to a corresponding free power, the truth or falsity of its parts is subject to divine will. (1.21, 231)
One of Bradwardine's examples of the absolute necessities is the conditional: "If this creature exists, God is its Lord." The truth of the conditional is prior to God's will, but the truth of the antecedent depends on God's will (1.21, 231). Anselm similarly thinks that there are necessary truths which do not depend on God's will in the sense that they would be freely chosen, for example those about God's existence and attributes and the principles of logic.34 As for the created world, Anselm states that whatever occurs takes place either by natural causes, the functions of which are determined by God, or by the acts of the free will, or by God's power and will. Events which take place in accordance with the common course of nature are naturally necessary or impossible with respect to the laws of the created order, but God as the Lord of this order can bring about effects which are naturally impossible. Miraculous divine interventions do not violate natural patterns, since they are meant to be subordinate.35 Anselm's view of the miracles is similar to that of Damian, though Anselm states that when something has happened it is no longer able not to have happened and it is always true that it has happened. According to Anselm, one should not say that God cannot change this truth or that God necessarily accepts it, for "God, since He is the truth, wills that the truth always be immutable, just as it is."36 Anselm's above remarks on God's will and modalities are based on the view that nothing is necessary or impossible to God in the proper sense of these terms, "necessarily" as referring to being compelled and "impossibly" to being prevented. All natural necessities and impossibilities of this kind are caused by the creator. I will return to Anselm's distinction between proper and improper modalities in the last section.
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila Damian's second concern pertaining to the alleged limitations of divine power was the deterministic interpretation of the Aristotelian principle that what is, necessarily is, when it is. In discussing questions pertaining to future contingents, foreknowledge, and freedom, Anselm deals with this dialectical question by applying the distinction between "antecedent" (necessitas antecedens, necessitas praecedens) and "subsequent" necessity (necessitas subsequens, necessitas sequens).37 Events or states of affairs compelled or constrained by external causes are necessary by antecedent necessity.
When something is called necessary in the sense of subsequent necessity, there is no reference to such a constraint but only to the fact that nothing can affect its being the case. Anselm says that this is the Aristotelian necessity by which all that has been necessarily has been, all that is necessarily is and necessarily is going to have been, and all that is going to be is necessarily going to be.38 Some of Anselm's formulations of the subsequent necessity do not differ from Boethius's temporal necessity. In Cur Deus Homo 2.17 Anselm writes:
When the heavens are said to revolve because it is necessary for them to revolve, this is the antecedent and efficient necessity. When I say that because you are speaking, you are necessarily speaking, this is the subsequent necessity which does not cause anything but is caused. For when I state this, I mean that nothing can bring it about that you do not speak while you are speaking. (S ii: 125.9–13)
The actuality of a thing excludes the possibility that it is not actual at that time. This necessity is said to be caused by actuality:
But this kind of necessity does not compel a state of affairs to occur; rather the existence of the state of affairs causes the existence of the necessity.
(S ii: 125.6–7) Some authors have characterized Anselm's distinction between antecedent and subsequent necessity as a distinction between physical and logical necessity.39 This is misleading, since no kind of logical necessity is dealt with in the statements above. They are clearly based on the traditional doctrine of the necessity of the present and its application to past and future things, which are also immutable in so far as they are past or future facts.
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
Anselm explains the temporal aspect of the subsequent necessity as follows in De Concordia 1.2. Events are past, present, or future with respect to a particular present moment of time. Insofar as they are contingent in themselves, they are not necessarily past, present, or future, but qua past events they are necessarily past, qua present necessarily present, and qua future necessarily future. As for contingent past events, before they occurred it was possible that they would not occur, but when they are past, it is not possible that they have not occurred since their antecedent possibility of not occurring is lost. The same applies to present events. Some of the future things have the possibility of not happening in the future.40 All this is in accordance with the traditional idea that alternative possibilities refer to the future. The necessity of the present and past is a corollary of treating possibilities as potencies with respect to actualization in one linear history. According to Anselm, a past event qua past is necessarily past, just as a white thing, qua white, is necessarily white.41 Formulations of this kind have contributed to attempts to read Anselm's subsequent necessity as the "logical" necessity "if p, then p."42 However, Anselm's accounts of this necessity show that he did not operate with any clear distinction between "logical" and "temporal" necessity. It is worth noticing that he sometimes associates the subsequent necessity with the form "Because x is, x is" and sometimes with "If x is, x is" without commenting on the difference.43
Anselm associates the subsequent necessity with the truth of statements about future contingent events. In explaining Mary's believing the truth of a prophetic statement about the death of Christ,
Anselm writes:
Therefore, since her faith was true faith it was necessary that things would be as she believed. But if you are once again disturbed by my saying "It was necessary . . . ," then remember that the truth of the virgin's faith was not the cause of his dying freely but that her faith was true faith because this was going to happen. (Cur Deus Homo 2.17; S ii: 124.27–125.3)
Anselm thinks that future contingent propositions are true or false and that this does not have an influence on the contingency of the events. This was suggested by Boethius, but Anselm formulated the idea more explicitly. A similar theory was employed by most scholastic thinkers.44 It is analogous to those contemporary
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila approaches in which "is" is regarded as tenseless when temporally definite propositions are said to be true or false. Each proposition has a truth-value, but the historical states of affairs, which are the truth-makers of future contingent propositions, are not yet causally fixed.45 Antecedent causal necessity can determine the truth-maker of a future proposition, but this does not happen to propositions which refer to the acts of free will. Such an act cannot be determined and, before the act, its alternatives are possible as well. In this context
Anselm makes use of diachronic modalities which were extensively analyzed by Boethius. In addition to antecedent necessities there are antecedent alternative possibilities of which the unrealized ones cease to be possibilities.46
In his De Grammatico Anselm employs two modal syllogisms which D. P. Henry regards as signs of fairly systematic discussion of the logical properties of such syllogisms.47 This is merely a speculative suggestion. Anselm makes use of some further modal principles as well, such as "from an impossibility something impossible follows," but he does not systematically discuss modal consequences or modal syllogisms.48 the elements of modal semantics The incomplete Lambeth Fragments involves preliminary notes for a work in which Anselm planned to present a general analysis of predication and, as part of it, a general account of the uses of modal terms. Although the project remained unaccomplished, some main features of Anselm's modal semantics can be formulated on the basis of his remarks in this brief text and in some other works. Anselm thought that a general theory of predication could be built on an analysis of the possible bases for the ascription of facere (to do, to make, to bring about). The combination of facere and predication was based on the view that in some sense a subject can be considered a cause of its predicate.49 Anselm thought that the causes are either efficient or nonefficient; allowing for nonefficient causation makes the "causal" theory of predication somewhat more understandable.50
Eileen Serene has summarized Anselm's analysis of direct and indirect modes of agency in the Lambeth Fragments as follows. A subject
(A) may be said to bring about a state of affairs (s) directly or through other states of affairs (m, n, o, r) causally related to s only if at least one of these six conditions is satisfied:
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006
Anselm on modality
(i) A directly brings about s; (ii) A directly fails to prevent s from occurring;
(iii) A brings about m, and m causally contributes to the occurrence of s; (iv) A fails to bring about n, and n's not occurring causally contributes to the occurrence of s;
(v) A prevents o from occurring, and o's not occurring causally contributes to the occurrence of s; (vi) A fails to prevent r from occurring, and r's occurring causally contributes to the occurrence of s.51
Anselm states that a verb is properly ascribed if and only if the subject directly brings about what is ascribed to it. In all other cases the ascription is improper.52 He assumed that the above distinction or some parts of it apply to all verbs and predicates and that analyzing the relationships between the proper and improper uses of terms was an important philosophical task. In a brief note he suggests that one could investigate the differences between proper and improper uses when a term is applied to a genus and its species, a cause and its effect or a whole and its part.53 Some contemporary thinkers have found Anselm's analysis of facere interesting, because it shows certain similarities to modern theories of the logic of action.54
In Anselm's view debere ("ought") and posse ("can") can be analyzed in the same way as facere.55 While there is a brief discussion of "ought" in the fragments, it does not involve an analysis of "can." What Anselm says about the proper meaning of necessity and possibility in his earlier works is pretty straightforward: "is necessarily something" properly ascribes a constraint to its subject, and "is possibly something" properly ascribes a capability or power to a subject. Modal terms in their proper senses refer to properties of things, such as the power to bring something about or received constraints. If modal terms are used in any another way, they are used improperly.56 In the Lambeth Fragments Anselm seems to regard the notion of capability as the basic modal notion, and there are examples of this way of thinking in other works: something is necessary in an improper sense with respect to the causes which cannot prevent it, and something is similarly impossible with respect to the causes which cannot bring it about.57 While Anselm gave many examples of improper uses, he did not develop a detailed theory of improper modalities. Clearly, however, Anselm was interested in Cambridge Companions Online © Cambridge University Press, 2006
Cambridge Companions Online © Cambridge University Press, 2006
simo knuuttila these because of their importance for certain theological doctrines.
God is necessarily truthful in the improper sense of necessity. While there is no constraining proper sense necessity in God, His truthfulness is necessary in the sense that there is nothing which could bring it about that He does not declare the truth.58 While God cannot do things which can be done only through impotence, there is no proper impossibility in the sense of inability in God.59 When it is said that "This man can be overcome," the possibility is not said properly, because of an ability in him, but improperly, because of an ability in someone else. Similarly "This man cannot be overcome" does not imply an inability in the man but in others.60 "The will of this man is overcome by temptation" is not true at all, since the will is only overcome by its own power.61 As stated above, Anselm was particularly interested in the question of how a thing can exist before its exists. On his analysis it can exist in the improper sense of "can," because something else can actualize it.
In discussing the possibility as an ability, Anselm thought, as did Aristotle, that full potencies or possibilities are combinations of various partial potencies. In Chapter 3 of the De Libertate Arbitrii,
Anselm analyzes one's ability to see a mountain; its components are said to be: (1) the power of seeing in one who sees, (2) the power in the thing to be seen, (3) the power which helps the sight (light), and (4) the power consisting in the fact that nothing obstructs the view. It is then stated that if one of these four powers is missing, the other three are not able to accomplish the result. A person can see the mountain in the full sense of the word only when the four powers are present.
In Metaphysics 5.12 and 9.1 Aristotle characterizes potency, the principle of motion or change, either as an activator or as a receptor of a relevant influence. The theory of active and passive potencies was meant to explain how and why a particular change takes place.62
This model allowed Aristotle to speak about all kinds of unrealized partial possibilities by referring to various levels of potentiality or by treating passive and active potencies separately. A full possibility requires that the active and passive factors are in contact and the actualization is not externally prevented, the possibility being necessarily realized in such a case (Met. 9.5 [1048a5–21]). This way of looking at possibilities is associated with the problem that there seems to be no difference between a full possibility and its actuality.
Cambridge Companions Online © Cambridge University Press, 2006 Cambridge Companions Online © Cambridge University Press, 2006