Journal of the History of Philosophy

introduction

immortality was an intensely debated topic in eighteenth-century German philosophy. Like most philosophers steeped in the Leibnizian tradition, Moses Mendelssohn attempted to prove the immortality of the human soul by proving, first, its indestructibility as a precondition for personal immortality proper. In the first of the three dialogues that make up his Phädon, he advances a complex argument to the effect that the soul cannot naturally cease to exist because simple beings can perish only by sudden annihilation or by gradual extinction, both of which involve a (naturally) impossible leap from being to nothingness. In the second edition of the Critique of Pure Reason, Kant included a "Refutation of Mendelssohn's Proof of the Persistence of the Soul" in the Paralogisms section (B 413–15), in which he argues that the soul might disappear by "gradual remission of all its powers" or "elanguescence" (B 414). The interpretation of both [End Page 75] Mendelssohn's proof and Kant's refutation is highly controversial.¹ In particular, there is no agreement as to whether Kant correctly construed Mendelssohn's proof,² nor as to whether he overlooked the fact that the Phädon does address the possibility of elanguescence or gradual extinction.³

Thanks to Kant's refutation, the Phädon's proof acquired visibility but was primarily read with an eye to either confirming or opposing Kant's verdict. Little effort was made to disentangle the proof from the Kantian context and explore its original background. Even when the eighteenth-century German context was taken into consideration, scholars found it difficult to go beyond the generic category of "Leibnizian-Wolffian philosophy," thereby assuming that Wolff's positions were in substantial agreement with Leibniz's.⁴ As we will see, this was not always the case.

This paper offers the first detailed investigation into the multiple direct sources of Mendelssohn's argument. At first sight, approaching the Phädon from this perspective seems at odds with Mendelssohn's attitude, since in the dialogues he cites no sources and anachronistically puts modern arguments "in the mouth of Socrates" (Phädon, appendix, 152). In the preface, on the other hand, Mendelssohn acknowledges his debt to ancient and early modern philosophers like "Plotinus, Descartes, Leibniz, Wolff, Baumgarten, Reimarus, et al.," thus inviting the reader to distinguish his borrowings from his own original contribution (Phädon, 43).⁵ Certainly, Mendelssohn excelled in borrowing arguments from other authors and recasting them for his own purposes. The argument under consideration is a case in point. Mendelssohn drew inspiration from Roger Boscovich's doctrine of continuity to build an argument whose premises were Boscovichian but whose overall framework and aims were Wolffian. The result was original in that neither Wolff nor Boscovich would have endorsed it.

Whereas knowledge of Wolff's legacy in the German Enlightenment has improved in the last few decades, Boscovich's influence is still largely underrated. For a long time, his role in the development of eighteenth-century thought was acknowledged (if at all) only in the fields of dynamics and physical monadology.⁶ [End Page 76] Although recent contributions have drawn attention to his doctrine of continuity,⁷ its reception outside mathematics and physics remains almost unexplored. Mendelssohn's interest in Boscovich's doctrine shows how relevant continuity was to metaphysics as well. More generally, this philosophical episode is emblematic of the links between mathematics, physics, and metaphysics in the pre-Kantian age.

Recently, Jonathan Simon and Colin Marshall have proposed a thought-provoking interpretation of both the Phädon's argument and Kant's refutation. Their approach shows typical features of rational reconstructionism: emphasis is placed upon the intrinsic philosophical interest of the exchange; the reconstruction of the arguments tends to maximize their soundness and transhistorical significance, which also allows comparison with arguments from contemporary analytic metaphysics; and, most importantly, the authors' aim is not so much to determine what arguments Mendelssohn and Kant actually advanced or intended to advance as it is to reconstruct the best arguments they were in a position to construct, given the assumptions they actually accepted.⁸ This does not mean that Simon and Marshall are insensitive to context or indifferent to whether their reconstruction would sound acceptable to Mendelssohn and Kant. Yet, their approach is not a contextualist one. Although they do not reject the principle that Christia Mercer dubs the "Getting Things Right Constraint,"⁹ their primary concern is not whether or not the arguments they attribute to Mendelssohn and Kant are ones that those philosophers would recognize as their own, but rather whether or not such arguments are ones that Mendelssohn and Kant could recognize as their own. Thus, they feel no obligation to track down any formulation of those arguments in the very texts they are interpreting. Far from criticizing this approach, I believe that Simon and Marshall have significantly improved our understanding of the Phädon's argument. The contextualist reconstruction I propose confirms that at least part of their interpretation is successful in that it helps us understand what was really at stake in Mendelssohn's concern with the final moment of existence, although the argument I ascribe to the Phädon is substantially different from their "mereotopological" argument. In spite of the differences in method and results, my own ecumenical wish is to suggest that both approaches can fruitfully stimulate, confirm, and correct each other.

2. wolff's doctrine of annihilation

During the 1720s and 1730s, the Leibnizian distinction between immortality and indestructibility became a cornerstone of Wolff's rational psychology. Along with his disciple Ludwig Philipp Thümmig, Wolff developed a thorough account of the specific meanings that words like 'death' and 'destruction' acquire when they [End Page 77] are transferred by analogy from the body to the immaterial soul.¹⁰ As concerns the living body, death is the moment in which all the organs "entirely cease from their functions."¹¹ After death, the body undergoes processes of corruption or putrefaction that eventually lead to its final destruction (interitus), "whereby the structure not only of the whole body but also of individual parts, however tiny, completely dissolves" (Thümmig, Demonstratio immortalitatis, §3). Wolff even characterizes death as "the state of the body from which its destruction follows" (Psychologia rationalis, §735). Bodies cease to exist when they disintegrate into parts, so that their physical identity is lost. However, their matter is not thereby annihilated. In the physical world, destruction does not entail annihilation (Wolff, Cosmologia generalis, §125).

The case of the soul is different, for its simplicity entails that the soul is immune to the process of corruption that affects the dead body and eventually causes its destruction: "The soul is a simple substance, devoid of parts; hence it is impossible for it to dissolve into parts and, when this dissolution is completed, to cease to exist" (Wolff, Psychologia rationalis, §729). However, incorruptibility is not sufficient to rule out that the soul is exposed to death and destruction; it only makes the transition from death to destruction necessarily different from what happens to the body. In the case of bodies, this transition is not immediate but gradual; there is a (shorter or longer) interval between physical death and destruction. By contrast, for incorruptible entities like souls or spirits, the transition must be immediate: "If a spirit is destroyed, it is destroyed in an instant"; its interitus must be "instantaneous" (§671), since it cannot be a consequence of gradual decomposition. In the case of simple substances, destruction can only consist in annihilation: "The soul can perish only by annihilation" (§732). And since destruction is the inevitable consequence of death, the death of the soul must result in its immediate annihilation. The death of the soul, unlike that of the body, leaves no cadaver behind: the soul must either live or cease to exist at all. If it dies, it must immediately vanish.

All the foregoing concepts and distinctions are echoed in the Phädon's first dialogue. Like Wolff, Mendelssohn distinguishes different possible terminations of life and existence: death, destruction, and annihilation. Mendelssohn's term Untergang is indeed the German rendition of the Latin interitus and—with reference to the body—denotes the complete dissolution of the corporeal machine into its tiniest parts, the final stage of the process of corruption. Like Wolff, Mendelssohn points out that the destruction of the body does not involve annihilation, for none of the smallest parts actually cease to exist: they simply disaggregate and recombine in a different way (Phädon, 93). What about the soul?

Kant's refutation begins by placing the Phädon's argument in the context of the traditional simplicity argument for the indestructibility of the soul:

This acute philosopher soon noticed that the usual argument through which it is to be proved that the soul (if one grants that it is a simple being) cannot cease through disintegration, is insufficient for the aim of securing the soul's necessary continuing duration, since one could still assume cessation of its existence by vanishing. In his [End Page 78] Phaedo, he sought to avoid this perishability, which would be a true annihilation.

(Critique of Pure Reason, B 413)¹²

Here, Kant implicitly suggests that Mendelssohn was the first to address the hypothesis that the soul might cease to exist by vanishing. Though accepted by scholarship,¹³ this picture is untenable. Both Leibniz and Wolff recognized that the indestructible simple being that is our soul can nevertheless be annihilated, but they ascribed to God alone the power to create or annihilate.¹⁴ Wolff advanced at least two arguments to the effect that annihilation, while metaphysically possible, is physically impossible, so that simple beings cannot be annihilated by natural causes. In his 1736 Theologia naturalis I, §861, the impossibility of natural creation and annihilation is established from the principle that only God has the power to conserve things in existence and thus also to annihilate them. This theological argument has little relevance to the Phädon, in which Mendelssohn tries to do without theological assumptions. More relevant is the cosmological argument sketched in the 1720 German Metaphysics, §102. First, Wolff recasts the traditional simplicity argument to the effect that simple beings cannot be destroyed like composite beings:

1. (1). If something has no parts, it cannot be modified by the separation or transposition of parts.

2. (2). If something cannot be modified by the separation or transposition of parts, it cannot cease to exist by the separation or transposition of parts.

3. (3). If something cannot cease to exist by the separation or transposition of parts, it can cease to exist only by annihilation.

4. (4). Simple beings have no parts.

5. (5). Thus, simple beings can only cease to exist by annihilation.

Second, Wolff implicitly applies to annihilation what he has established about creation, namely that simple beings cannot be formed little by little, since they contain no plurality of parts that may follow one another. If their existence is to begin, it must begin at once. Thus, creation cannot be gradual and does not happen over time but in an instant (§§89, 101). For the same reason, annihilation cannot be a gradual, temporal process:

1. (6). If a simple being is to cease to exist, it must be annihilated at a single blow.

Although this conclusion about annihilation is left implicit, it clearly serves as a premise in the third and final section of the argument, which is developed in the cosmological chapter of the work. Here, Wolff compares natural events with supernatural events in terms of their respective temporal structures and finds that both creation and annihilation belong to the latter group. His proof begins with the premise that gradualness is the hallmark of natural events and with the characterization of miracles as nonnatural (i.e. supernatural) events:

1. (7). For every event e, e is natural if and only if it happens gradually (§687).

2. (8). For every event e, e is natural if and only if it is not a miracle (§§633, 688). [End Page 79]

3. (9). Hence, for every event e, e is a miracle if and only if it does not happen gradually (§688).

Miracles lack precisely the gradual character of natural events; they do not happen little by little over time but all at once, in single instants of time. Drawing on his former conclusions about creation and annihilation, Wolff further concludes that

1. (10). "Simple beings cannot begin or cease otherwise than by miracle, insofar as they must begin and cease at once and in an instant, if they are to begin and cease at all." (§688)

Most interesting for our purposes is how Wolff justifies and elucidates premise (7). A staunch mechanist, he argues that since all changes in the physical world are brought about by the motions of bodies or parts of bodies, and since the division of matter makes the communication of motion happen little by little, "all changes in bodies must happen little by little as well; thus, all events must arrive little by little, by degrees" (§686). Wolff considers this proposition to be equivalent to the traditional adage concerning the impossibility of leaps in nature: "And this is what is meant by saying that nature makes no leap" (§686).

Thus, long before Mendelssohn, the German Metaphysics had emphasized the incompatibility between the punctual, instantaneous character of annihilation and the continuous, temporal character of natural events. The annihilation of a simple being cannot be the work of nature insofar as it involves a leap. The continuity of natural processes is invoked to relegate all discontinuous events either to the supernatural dimension or to the world of dreams and fairy-tales, in which events happen that do not conform to the Principle of Sufficient Reason (henceforth, PSR). Like other Leibnizians who attempted to derive the Law of Continuity from the PSR (see below), Wolff understands the absence of leaps primarily as a manifestation of the rational connection of things in the world. To avoid misunderstandings concerning leaps, he points out that to happen by leaps is not simply to take place "suddenly," in the merely temporal sense of "very quickly": "Actually, the leap does not refer to time but to the connection of beings, to how they come from one another. And this happens by degrees, if in the preceding being we always find a sufficient reason why the other comes from it" (German Metaphysics, §698). The ignition of gunpowder, for instance, seems to be a sudden process because of how quickly motions propagate from a spark, but it nevertheless happens step by step (§691).

There is a leap in a series of states if and only if some previous state contains no sufficient reason for the following state: "Thus, the sufficient reason hinders leaps, just as its absence in dreams can tolerate them" (§689). The necessarily gradual character of natural processes is but a consequence of the PSR.¹⁵ For instance, if a body is to move from one place to another, it must go "through all the intermediate places," since nature makes no leap. But if someone further asks "why nature tolerates no leap here, then we must answer as in all other cases in [End Page 80] which this question is raised, namely: because otherwise it would be impossible to conceive how the body came from the one place to the other, and therefore this change of place would have no sufficient reason" (§690). Leaps cannot take place in nature because they are unintelligible and would make events impossible to explain from their causes.

3. mendelssohn's argument against annihilation

The idea that nature always avoids discontinuities is pivotal to the Phädon's argument. In Mendelssohn's eyes, this is indeed the main modern addition to the traditional Platonic argument. Only the doctrine of continuity "provides a high degree of certainty" in these matters and "leads us to proper conceptions of the changes of the body and the soul, without which one cannot consider death and life, mortality and immortality, from the proper view point" (Phädon, appendix, 151). In particular, the principle that nature makes no leap inspires Mendelssohn's attempt to establish, from the impossibility of discontinuous changes in nature, the impossibility of annihilation or transition from existence to nonexistence.¹⁶ In the first dialogue, Socrates asks what happens when the structure of the body is entirely decomposed into its smallest parts:

Do these parts cease to undergo changes? Are they entirely lost? . . . Impossible . . . for is there a mean between being and non-being? . . . Therefore being and non-being would be two states which immediately follow on one another, which must be the closest [i.e. contiguous] to each other: however we have seen that nature can produce no such changes, which happen suddenly and without transition.

(93)

Thus, sudden annihilation cannot be an option even for the soul. Mendelssohn's Socrates considers the alternative hypothesis—the soul disappears by the gradual yet complete extinction of its powers¹⁷—but rejects it for the same reason he rejects sudden annihilation. However gradual, the loss of powers cannot result in complete extinction otherwise than by a sudden leap from existence to nonexistence:

The soul cannot perish in eternity; for the final step—one may postpone it as long as possible—would still always be a leap from being to nothingness, which can be grounded [i.e. have a sufficient reason] neither in the essence of an individual thing, nor in the entire connection [of things]. Therefore the soul will continue and be eternally existent.

(98, translation modified)

To deny that there can be a sufficient reason for the leap from being to nothingness either in the essence of the thing that disappears or in the universal connection of things is to deny that nature can make such a leap. Of course, Mendelssohn's recourse to both the Law of Continuity and the PSR to rule out natural annihilation has a Wolffian flavor. However, Wolff's influence alone is sufficient neither to account for Mendelssohn's unfaltering reliance on the Law of Continuity nor to provide a satisfying reconstruction of his argument. In particular, Wolff did not [End Page 81] offer a suitable concept of the continuum. Instead of characterizing the continuum in terms of infinite divisibility or density, he characterizes it as a being whose parts "follow one another in their order in such a way that no other parts can be put in between them in a different order" (German Metaphysics, §58). This account of continuity as an inalterable order became standard. As late as 1740, it appears, for instance, in Du Châtelet's Institutions de physique (101–2). Conceiving continuity without infinite divisibility was consistent with the idea that change is gradual—all intermediate places are traversed, all intermediate values are successively taken on—but nevertheless discrete, as the number of intermediate degrees to traverse is finite.¹⁸ In this context, Maupertuis's 1750 attack on the Law of Continuity seemed to pose a serious challenge: however gradual, the least change is as discontinuous as "the sudden destruction of the universe," insofar as it involves a transition from one degree to another (Essay de Cosmologie, 67). Of course, it was possible to counter the objection by invoking the PSR to discriminate between gradual changes and sudden annihilations. Gottfried Ploucquet argued that the transition from one degree to another would be a leap only "if the latter [degree] did not have its fundament in the former" (Dissertatio historico-cosmologica, 51). However, he himself admitted that continuity requires infinite divisibility, without which it turns into discrete quantity (see 52); and as long as degrees were conceived of as discrete values on a scale, it seemed difficult to escape Maupertuis's point that even the minimal transition from one degree to its immediate successor must involve a leap, however small. Thus, what Mendelssohn needed was a doctrine of continuity immune to Maupertuis's objection. He found it in the works of Roger Boscovich, whose express aim was to counter Maupertuis's attack and restore the Law of Continuity as a universal law of nature.¹⁹

Before exploring Mendelssohn's direct source, it is useful to take a look at the critical reactions to his argument. According to Jonathan Bennett, the Phädon's argument fails because it wrongly assumes that discontinuous change would entail that something is in a certain state at time t and in a different state at the next instant after t; but "there is no 'next instant,' for the set of instants-laterthan-t has no earliest member."²⁰ More charitably, Simon and Marshall initially base Mendelssohn's argument on the Law of Continuity, which "rules out a natural process of change that can be represented by a discontinuous function defined over a continuous time parameter."²¹ However, they deem continuity insufficient to forestall the possibility—later exploited by Kant—of a protracted elanguescence that ends up in the being's utter vanishing. They contend that in the case of elanguescence one is not justified to appeal to the Law of Continuity. Mendelssohn's point against disappearance by elanguescence "is that such a process must involve a specific final moment (a 'leap' from being to nothingness), [End Page 82] not that it must skip over some intermediate values (of power),"²² since the very process of elanguescence is assumed to exhaust all such intermediate decreasing values. As a consequence, elanguescence may involve a transition from being to nothingness without violating the Law of Continuity.

This is actually Kant's point both in the first Critique and elsewhere. His lectures on metaphysics from 1782–83 feature an early refutation of Mendelssohn's proof:²³

Mendelssohn attempted to prove the immortality of the soul as substance first, but he did not succeed. He says: it is simple, therefore cannot be decomposed through division. But it can expire and its reality gradually diminish. He says that would be a leap: but it could still happen by the law of continuity [Er sagt, das wäre ein Sprung; aber es könnte ia noch beim Lege continui geschehen].

(Metaphysik Mrongovius, AA 29:912/ Lectures on Metaphysics, 277–78)

This passage makes it implausible that Kant simply overlooked the part of the Phädon's argument addressing gradual extinction, for Kant here correctly (though succinctly) acknowledges that Mendelssohn does have a response to this alternative: "He says that would be a leap." Kant seems aware that their real disagreement is as to whether this gradual expiration would violate the Law of Continuity or not. Yet, the reasons behind this disagreement are all the less clear as the two authors agree on several points. Both assume that the soul's powers are intensive quantities, which have degrees instead of parts.²⁴ Both assume that natural changes in both extensive and intensive quantities conform to the Law of Continuity.²⁵ Both consider the hypothesis of gradual diminution and final extinction of the soul's powers. However, they disagree as to what happens at the end of the process. Mendelssohn maintains that the final step should be a leap from being to nothingness, whereas according to Kant there would be no leap, for the soul's powers would decrease to zero without skipping any intermediate degree.²⁶

To explain Mendelssohn's attitude to elanguescence, Simon and Marshall propose a sophisticated reconstruction, one that does not base his argument on the Law of Continuity but on a corollary of the PSR, the Principle of Mereotopological [End Page 83] Sufficient Reason. I address this proposal in section 5, after advancing my own interpretation, which by contrast insists on the role of the Law of Continuity in its Boscovichian formulation (sections 4–5). Finally, I propose the hypothesis that Mendelssohn's position concerning gradual extinction was the outcome of his Wolffian approach to Boscovich (sections 6–7).

4. boscovich on natural changes

In the Phädon's second edition, Mendelssohn discloses the source of his doctrine of continuity: "Father Boscovich, who has put the Law of Continuity in an excellent light, deserves to be read again on this subject" (Phädon, appendix, 151). In a footnote, he refers to Boscovich's De continuitatis lege (1754) and Philosophiae naturalis theoria (1st ed. 1758), the latter of which he had reviewed in 1759 for a German literary journal.²⁷ The impact of both works on the Phädon's first dialogue (composed between 1761 and 1763) appears considerable. Although Mendelssohn's reference to Boscovich is regularly noted in the literature,²⁸ this source appears to be still underexplored; little has been done to assess whether and how the adoption of the specifically Boscovichian idea of continuity affects the Phädon's argument.²⁹

Boscovich's concept of continuity includes both continuity in change, which rules out gaps between successive values, and spatiotemporal density, which rules out perfect proximity between any two points or instants. In Boscovich's formulation, the Law of Continuity concerns changes in quantities. It states that "any quantity, in passing from one magnitude to another, must pass through all intermediate magnitudes of the same class" (Theory, §32), which entails that no increase or decrease in a quantity can happen in an instant, but only in the interval of time between two instants. This is key to understanding why continuous changes include no gap between successive values. The distinction between instants and intervals of time falls within the general distinction between boundaries and parts that applies to the continuum as such, thereby making it possible to represent time and change by means of geometric relations. Citing Aristotle,³⁰ Boscovich maintains that the very nature of the continuum entails that any two consecutive parts of it must have a common boundary. Two parts of a line are delimited by a point, two parts of a surface by a line, and two parts of time by an instant. Likewise, two consecutive variations of a quantity are delimited by a determinate [End Page 84] state or magnitude of that quantity. Instants of time and states of quantities can be represented by geometric points because, just like points, they are only boundaries between consecutive parts in a continuum.

Boscovich offers both an inductive and a metaphysical (i.e. a priori) proof of his law. The latter is of major interest to us. In a Newtonian fashion, the continuous variation of a quantity is geometrically represented by the "flux" of an ordinate that, shifting along the horizontal axis that represents the continuous succession of time, acquires some determinate magnitude at each single point or instant, so that there is no point in the axis or instant in time that lacks a corresponding magnitude of the ordinate, and vice versa (Theory, §34). Boscovich relies on such geometric representations³¹ of variation over time to argue that variation can only take place in continuous intervals (tempuscula), no matter how small, and not in instants (momenta temporis; see §33). His metaphysical argument pivots on the impossibility of contiguity between indivisible entities. Spatiotemporal density forbids immediate contact between any two points or instants. No two points can be perfectly contiguous to one another, that is, so close that no intermediate point can be found in between; on the contrary, there must always be a line that stretches from the one to the other, for if their distance were reduced to zero, they would necessarily coincide with one another and thus coalesce into one single point: "Nor are there two contiguous points, one of which is the end of the first segment, and the other the beginning of the next; for two indivisible and unextended things cannot be contiguous without their compenetration and coalescence into one" (§48, translation modified).

The underpinnings of this key principle affirming the identity of contiguous indivisibles come more fully to light in the 1754 dissertation on continuity. Here, Boscovich explains that the principle can be inferred from the very concept of indivisibility: "Indivisible things either are distant from one another or, if distance is removed, they coalesce into one. For, things that completely lack extension either do not touch one another or touch one another in their entirety [secundum se tota]. In the former case, they are distant from one another. In the latter, they compenetrate and coalesce into one" (De continuitatis lege, §10). The idea is that there cannot be any partial contact (such as contiguity) between things that have no parts. The only conceivable spatial relations between indivisible things are either full contact (i.e. coalescence) or distance. Thus, Boscovich's principle does not rest upon the imagination. On the contrary, our imagination is what misleads us into thinking that two points can touch each other at their margins, as though unextended things could have margins. Rather, the principle rests upon the utter inconceivability of contiguity without extension: "In order to conceive of a point as contiguous to another point yet located outside it and not compenetrated, one must do violence to oneself. One will conceive of some extended little globes which are in contact in one part but separated in another, whereby one will admit parts in the same point and destroy the idea of indivisibility and non-extension [inextensio]" (De continuitatis lege, §10). The same argument, Boscovich points out, has always been used to refute Zeno's account of the continuum as composed of unextended points. [End Page 85]

The principle also holds for both instants of time and states of quantities:

In the same way, there is no instant of time that is so near to another instant that has gone before it, that it is the next after it; but either they are the same instant, or there lies between them a continuous interval that can be infinitely divided by other intermediate instants. Likewise, there is no state of a continuously varying quantity so very near to a preceding state that it is the first state after it, some momentary addition having been made; but the difference between such states is due to the intermediate continuous interval of time.

(Theory, §36, translation modified)

From this impossibility of contiguity, Boscovich derives the impossibility of leaps in continuous changes. For the sake of brevity, let us restrict our consideration to the easiest cases, namely changes involving quantities that can neither have two magnitudes at the same time nor lack any magnitude at any instant—unlike, for instance, geometrical curves, which can have two or more ordinates corresponding to one single point on the horizontal axis, or none at all. Changes in the real world are of the former kind. In nature, if a quantity were to leap instantaneously from one magnitude to another without passing through all its intermediate states, it would have to acquire two magnitudes at the same instant, "namely, the last magnitude of the preceding series and the first of the subsequent series" (§49, translation modified). As these two different states cannot be contiguous in time, they must either be separated by an interval or be coincident in a single instant. The latter is impossible, given the assumption that quantities must have exactly one value at each instant:³² "Thus, in any finite real series of states, there must of necessity be a first term and a last; and so if a leap is made . . . there must be at the instant, at which the leap is said to be accomplished, a twofold state at one and the same time. Now, since this cannot obtain, that leap absolutely cannot obtain either" (§51, translation modified). Imagine some material whose density changes instantaneously in such a way that, for one hour, density is d and, for the subsequent hour, it is twice d. At the boundary between the two intervals of time, density should leap from one state to another. Thus, "in that instant of time which separates the two hours, there would have to be two densities at one and the same time, the simple and the double, and these are real terms of two real series" (§52).

5. boscovich in the phädon

Boscovich's treatment of continuity provides the crucial premises of Mendelssohn's argument, those concerning the continuity of change in nature, the continuity of time, and the perfect correspondence between the series of changes and the series of times. Concerning natural changes, Mendelssohn claims that "change does not happen suddenly, but gradually" (Phädon, 92), insofar as the transition from one state to the opposite one (e.g. from life to death) cannot be immediate but always requires a transition through intermediate states: "nature in all its changes knows to find an intermediate state, which serves as a transition, likewise, to go from one state to the opposite" (88); "nature must take all intermediate states with it, when it wants to alternate a state with its contrary" (89). To find evidence that this idea of change comes from Boscovich, one may compare the Phädon with [End Page 86] Mendelssohn's review of the Theory of Natural Philosophy. In the Phädon, for instance, the dissolution of the body machine is described as a process that "happens so gradually, in such a continuous order, that every state can be called a common boundary [gemeinschaftliche Grenze] of the previous and the following state" (96, translation modified). In his review of the Theory of Natural Philosophy, Mendelssohn explains that "Boscovich considers the state that pertains to each instant only as the common boundary [gemeinschaftliche Grenze] between the previous and the following magnitude" (JubA 5.1:58, my translation). In his dialogue, Mendelssohn uses the same vocabulary (with the same meaning) that he had used to expound Boscovich's doctrine of continuity.

Time is continuous in the sense that it is impossible to find two instants "which are the closest to each other," that is, "so close together that there is nothing between them" (Phädon, 90–91). Mendelssohn's rejection of the idea that time is made up of contiguous instants is reminiscent of Boscovich's attempt to infer continuity from the impossibility of contiguity. This debt fully emerges in the second edition of the work,³³ which consistently draws on Boscovich's terminological and conceptual distinctions between parts of time (or intervals) and boundaries (or instants):

The parts of time . . . have common boundaries. . . . The smallest portion of time is such a series of moments, and may be subdivided into even smaller parts, which still always preserve all the properties of time. . . . There are, therefore, no two moments that are the closest to each other, and between which it is impossible to conceive of a third.

(JubA 3.1:364/Phaedon; or, the Death of Socrates, 60–61, translation modified)

Reading Mendelssohn along Boscovichian lines confirms what Simon and Marshall point out against Bennett's reconstruction,³⁴ namely that Mendelssohn's "moments" are not parts of time but limits, boundary points in time, just as states are boundaries in the continuum of change.

Finally, from these premises concerning change and time, along with the assumption that "changes progress at the same pace as time," Mendelssohn derives the proposition that change is continuous on a par with time, for "there are no two states, which are the closest to one other" (Phädon, 91, translation modified)—two states "between which no further third state is to be found" (JubA 3.1:365, my translation). Like Boscovich, Mendelssohn models the structure of change on the structure of time: "The succession of the changes coheres with the succession of time, and is therefore so constant, so interdependent, that one can specify no states which are closest to one another, or between which no transition should take place" (Phädon, 92, translation modified). If an entity e changes over time, [End Page 87] for every instant t there is a corresponding state of e, but e's state does not change at t, for changes necessarily happen during continuous intervals,³⁵ the parts of which are never instants but smaller intervals:

Nature . . . changes things gradually, and in a continued series, one after another. The smallest part of this series is itself a series of changes; and however close the succession of one state may seem upon another [man mag zween Zustände so dicht an einander setzen, als man will], there is always a transition between them which links them together, as if it showed nature the way from the one to the other.

(JubA 3.1:65, 365/Phaedon; or, the Death of Socrates, 62, translation modified; see also Phädon, appendix, 150)

Thus, Mendelssohn's premises reflect Boscovich's doctrine of continuity. Even changes that appear discontinuous to our senses—as though bodies were leaping from one state to another—are in fact continuous sequences. To illustrate this form of sense deception, Mendelssohn observes, like Boscovich, that we perceive and imagine the days of the week as discrete units, whereas they actually form a continuum (see Phädon, 94; and Boscovich, Theory, §45). Mendelssohn exploits this point to argue that even the body's transition from life to death must conform to the Law of Continuity, in spite of the common impression that death arrives at one salient point in time: "There is no precise moment, where one could say: Now the animal dies. . . . Certainly the changes must seem separated to our senses, since they become noticeable to us only after a fairly long interval of time; but we know enough that they cannot be so in fact" (Phädon, 93–94). This passage is reminiscent of Leibniz's claim that "no one can specify the true time of death, which for a long time may pass for a simple suspension of noticeable actions, and is basically never anything else in simple animals" (Philosophical Essays, 141). Through this continuist approach to living processes, the Phädon partially revives Leibniz's idea that the event called 'death' is not the real termination of life.³⁶

However, Mendelssohn's most spectacular use of the Law of Continuity is his argument against annihilation. The first premise is an instance of the Principle of the Excluded Middle: there is no "mean between being and non-being" (Phädon, 93). Mendelssohn aims to show that, by virtue of this premise, annihilation proves incompatible with the Law of Continuity; and since nature never violates continuity, there can be no natural transition from existence to nonexistence. If we read the Phädon's passage again in light of the Boscovichian doctrines outlined above, it turns out that the crucial step concerns the impossibility of contiguity: "Therefore being and non-being would be two states which immediately follow on one another, which must be the closest to each other: however we have seen that nature can produce no such changes, which happen suddenly and without transition" (93).

States are like points in a continuum. States, points, and instants are not parts of the continuum but its boundaries. Two different states cannot be contiguous [End Page 88] in time because, if there were no distance (no time interval) at all between them, they would simply collapse into one. Two boundaries that immediately follow upon one another are one and the same boundary. This is precisely what happens to states of being and nonbeing if they are assumed to immediately follow upon one another. As there can be no intermediate state between being and nonbeing, the last instant of existence and the first instant of nonexistence should be contiguous. But since two instants in continuous time cannot be contiguous without coinciding altogether, the last instant of existence would be the first instant of nonexistence. An entity's transition from the state of being to the state of nonbeing should be instantaneous, in the sense that there would be an instant at which that entity would both be and not be—a blatant contradiction. Thus, Bennett's reconstruction does no justice to Mendelssohn, who perfectly agrees that there is no next instant—not merely because time is dense, but because indivisible instants cannot be contiguous. If my Boscovichian reading is sound, Mendelssohn's point is precisely that contiguity is ruled out by the continuity of time, whereby two points in time that lack intermediate points cannot but be one and the same point.

How did Mendelssohn get the idea of applying Boscovich's doctrine of continuity to the issue of annihilation? The idea actually came from Boscovich himself. After developing his argument against leaps in continuous and especially natural changes, Boscovich observes that this argument seems to entail the impossibility of both creation and annihilation. He introduces the issue as an objection against the argument itself:

Against this argument it would seem at first sight that there is something ready to hand which overthrows it altogether; whilst as a matter of fact it is peculiarly fitted to clarify it. It seems that from this argument it follows that both the creation of any thing, and its destruction, are impossible. For, if the last term of a series that precedes is to be connected with the first term of the series that follows, then in the very transition from non-existence [non esse] to existence [esse], or vice versa, it will be necessary that the two are connected together; and then at one and the same time the same thing will both exist and not exist, which is absurd.

(Theory, §52, translation modified)

This passage contains a plausible antecedent for Mendelssohn's argument against annihilation. Later in the Phädon's first dialogue, Mendelssohn reasons along the same lines to address the hypothesis of the gradual extinction of the soul's powers. However gradual, this process must have a nongradual "final step," a "leap from being to nothingness" (Phädon, 98, translation modified), which would make these two states impossibly contiguous. If the soul entity were to eventually cease to exist, the previous interval of its existence and the subsequent interval of its nonexistence would have to share a single boundary point in time, for there can be no intermediate state between them. Hence, the soul's state at such a boundary point would have to be both a state of existence and a state of nonexistence. Mendelssohn's point is not merely that a leap from being to nothingness would cause a discontinuity in the process, as though the decreasing amount of the soul's powers skipped over some values; rather, his point is that such a leap would cause an impossible contiguity of states. [End Page 89]

6. continuity or sufficient reason?

My Boscovichian reading of Mendelssohn's argument has significant similarities with Simon and Marshall's Arbitrary Mereotopology Argument. They assume that "a simple being's existence is bounded by a single point of time, so that it exists at all previous times and does not exist at any later times."³⁷ Then they argue that the boundary state itself cannot be characterized as a state of either existence or nonexistence without arbitrariness.

Both the Boscovichian and the mereotopological reconstruction focus on the boundary point between existence and nonexistence and exploit the impossibility of ascribing it to both intervals. In addition, the mereotopological reconstruction also considers whether the boundary could be ascribed to only one (or none) of the intervals and rejects this possibility by invoking the PSR (or a qualified "mereotopological" version of it). This part of Simon and Marshall's argument is theoretically impeccable but lacks textual support. Although Mendelssohn certainly endorses the PSR, he does not use it to rule out the specific condition of arbitrariness described by Simon and Marshall.³⁸ By contrast, in Mendelssohn's source we have found a clear rejection of leaps for the express reason that they involve a "twofold state" and therefore a contradiction: at the instant at which a leap is made there must be "a twofold state at one and the same time" (Theory, §51). Thus, I am inclined to think that Mendelssohn's intended argument against gradual extinction relies on considerations about continuity and noncontradiction rather than sufficient reason.³⁹ Boscovich's above-described corollary about annihilation, along with Mendelssohn's express commitment to Boscovich's doctrine of continuity, strongly suggests that the argument actually endorsed by Mendelssohn is closer to the Boscovichian argument than the mereotopological one. This is all the more the case as the Boscovichian aversion to leaps in nature is as successful as any aversion to arbitrary boundaries in explaining Mendelssohn's emphasis on the soul's final step being an impossible leap out of existence.

One might object that the Law of Continuity is historically and theoretically connected with the PSR. But although this may be true for Leibniz,⁴⁰ it is certainly not true for Boscovich, who rejects the Leibnizian PSR as being "entirely false," "pernicious," and useless (De continuitatis lege, §125). He opposes the PSR as incompatible with free will (§126) and refutes the very argument by which "the Leibnizians" sought to derive the Law of Continuity from the PSR.⁴¹ The apparent [End Page 90] lack of a cause or reason, argues Boscovich, is no guarantee that the effect does not take place (§§128–30). If Mendelssohn's argument is borrowed from Boscovich, it cannot rely on the PSR.

7. zero versus nothingness

As hinted above, Boscovich presents the argument against creation and annihilation as an alleged consequence of his argument against leaps in nature and therefore as an objection to it. If the doctrine of continuity entails the impossibility of creation and annihilation, this is for him a good reason for questioning the doctrine. Fearing that his argument might be used to challenge the creationist dogma, the Jesuit does not consider its usefulness for defending the indestructibility of the soul. In this respect, Mendelssohn's appropriation of the argument against annihilation appears entirely original.

In the case of a thing's transition from nonbeing to being or vice versa, there would be a conjunction of two series: the series of the thing's states of being and the series of its states of nonbeing. The former would begin or end at the exact point in time at which the latter would end or begin, respectively. Thus, at that boundary point shared by both series, the thing would both be and not be, which is impossible. To defeat this argument, Boscovich highlights its implicit precondition that both series be series of real states. If they are, the argument works: "Hence, if to one series of real states there succeeds another series of real states also, which is not connected with it by a common term, then indeed there must be two states at the same instant, namely those which are their two limits" (Theory, §52). But if the states of either series are not real, the conclusion does not follow. This makes it possible to resist the anticreationist argument by rejecting its presupposition that states of nonexistence are real. As there are no real states before creation and after annihilation, such events entail no impossibility:

Since non-existence [non esse] is mere nothingness, a series of this kind requires no extreme limit, but is immediately and directly cut off by existence [esse] itself. Wherefore, at the first and at the last instant of that continuous time, during which the thing exists, it will certainly exist, without this existence being simultaneously connected with non-existence.

(§52, translation modified; see also §55)

From the Scholastic adage that nothingness has no properties, Boscovich infers that nonexistence has no real boundaries. There is neither a first nor a last instant of nonexistence. Any contradiction arising at the boundaries is prevented by expelling nonbeing from the boundary points themselves, so as to avoid any contact between the series of states of existence and the series of states of nonexistence.

This refutation accounts for the possibility of sudden annihilation in a way that conflicts with Mendelssohn's reasoning. Whereas Boscovich denies that states of nonbeing are real states on a par with states of being, Mendelssohn appears committed to the opposite view, since he claims that, in the case of annihilation, "being and non-being would be two states which immediately follow on one another" (Phädon, 93). However, Boscovich is not in the least concerned with the immortality of the soul. His aim is to safeguard the compatibility between his doctrine of continuity and creationism. The annihilation he admits as possible [End Page 91] need not be continuous like natural changes: "In creation and annihilation, a quantity can be produced or destroyed with a finite magnitude" (Theory, §57).⁴² We might even say that annihilation cannot be continuous, for nothingness cannot be the final term of the process. If so, elanguescence is impossible, for the final point of a real series cannot be nothingness. That is why Boscovich draws a sharp distinction between the genuine annihilation of quantities and their reduction to zero: "It seems that sometimes the last term of a real series is nothing. But if we go deeper into the matter, we find that it is not in reality nothing, but some state that is also real and of the same kind as those that precede it, though designated by another name" (§58).

As physical quantities cannot increase to infinity, the curves that represent their variations can transition from positive to negative values and vice-versa only by passing "through the value nothing [per nihilum]," that is, zero. This zero value "is not really nothingness in itself, but a certain real state; and it may be considered nothing only in a certain respect" (§37, translation modified). A quantity whose magnitude is zero is not thereby annihilated. States like rest, zero velocity, or zero force are real states "very different from the true non-being" (§61). This is paramount to Boscovich's unified account of motion, gravitation, and collision. His diagrams represent the magnitudes of attraction as the positive values of the force curve and the magnitudes of repulsion as its negative values. The point at which the force inverts its sign and converts from attractive to repulsive or vice versa—the point at which the curve cuts the x-axis of distances—is the point at which the force intensity—represented on the y-axis—is zero.

8. transformation as annihilation

We may thus ascribe the following theses to Boscovich:

1. (1). Discontinuities or leaps in natural changes are impossible.

2. (2). Nevertheless, annihilation involves no real leap and is therefore possible (though not in the form of gradual extinction).

3. (3). Forces or powers can continuously decrease to zero without turning into nothingness or nonbeing.

Mendelssohn appropriates (1) and rejects (2), but what about (3)? If the forces or powers of the soul can—like physical forces—be reduced to zero without ceasing to exist, one might use (3) to counter the hypothesis of disappearance by elanguescence. However, the Phädon's treatment of this hypothesis appears incompatible with applying (3) to the soul's powers so as to rule out its gradual extinction. When Socrates asks whether the postmortem decay of body and soul would eventually result in the complete disappearance of all cognitive and volitional activities—which would plausibly correspond to the zero degree of the soul's powers—Cebes dismisses this scenario as "impossible" because such an outcome would amount to "a total annihilation, and no annihilation . . . exists in the ability of nature" (Phädon, 98). A soul entirely deprived of its power cannot subsist even for a moment. [End Page 92]

Despite Simon and Marshall's claim that "Mendelssohn appears to allow that souls can momentarily exist without power (or action or suffering),"⁴³ the two passages they quote provide no evidence for this; and the second one—if correctly translated—actually provides evidence of the contrary. The first passage is Mendelssohn's admission that "a total deprivation of all consciousness . . . would not be totally impossible, at least for a short time" (126). What Mendelssohn says about conscious thought cannot be extended to the whole perceptual activity. For a Leibnizian like Mendelssohn, perception does not require apperception. "Sleep, fainting, dizziness, ecstasies" (126) are states of obscure or confused perception, not of total inertia.⁴⁴

The second passage is taken from the appendix to the Phädon's third edition. Here, Mendelssohn admits that the specific cognitive faculties of the soul (reflection,⁴⁵ inventiveness, judgment, etc.) can be "entirely inactive for a while," but—in the wake of Wolff—he carefully distinguishes those faculties from the "original power" that is their source (JubA 3.1:145, my translation). Even when certain mental faculties are inactive (as when the person is sleeping, distracted, beguiled, etc.), the soul's power "is anything but inactive [nichts weniger als unthätig]."⁴⁶ At no instant is the soul entirely inert or powerless. If its power were completely extinguished for a moment, the soul would not exist at that moment. That is why Mendelssohn's Socrates concludes that the soul cannot completely cease to produce representations, perceptions, thoughts, and so on; for otherwise it would vanish into nothingness, which is impossible. A "power to think," confirms the Abhandlung von der Unkörperlichkeit der menschlichen Seele, "must either think actually [würcklich dencken] or cease to exist" (JubA 3.1:175).

What is crucial in this context is the distinction between the soul's essential powers, which perform "fundamental activities" like thinking and willing, and its derived faculties, whose performances are "mere modifications of other activities" (Phädon, appendix, 150).⁴⁷ The latter conform to Boscovich's account of the variation of forces. The intensity of a sensation, for instance, can decrease to zero without there being a case of annihilation; its gradual extinction still falls within the realm of natural change.⁴⁸ By contrast, the disruption of the soul's [End Page 93] fundamental activities cannot be a natural effect: "All natural powers can only change determinations, only alternate modifications with one other, but they can never transform fundamental attributes and self-subsistent activities into nothingness; therefore the power (or powers) to think and will can never be annihilated by natural changes" (appendix, 150, translation modified).

Why does the power to think and will enjoy such a special status? In the Wolffian tradition, life and death were characterized in functional terms. To function is to live; to cease to function is to die. Both Thümmig and Wolff had tried to explain "in what sense" the soul could be mortal by claiming that the death of the soul should be conceived, by analogy with the death of the body, as the complete cessation of the soul's functions, which would necessarily result in the soul's annihilation: "The soul would be mortal if the exercise of all faculties ceased simultaneously and thence followed the annihilation of the soul" (Wolff, Psychologia rationalis, §736). The soul would be mortal if all its perceptions and appetitions ceased at some point in time and this death involved an utter destruction, which would leave nothing identifiable as a soul (see Thümmig, Demonstratio immortalitatis, §4). Death in this strong sense of the word cannot be confused with mere change or the transition of an entity to a different state.

This Wolffian background may also help explain why Mendelssohn is committed to the view that even the gradual extinction of the soul's power would have to involve a final leap from being to nothingness, or why he assumes that Boscovich's argument against leaps in nature applies to the case of elanguescence as well. My suggestion is that Mendelssohn approaches the hypothesis of gradual extinction from a Wolffian point of view. At first sight, Wolff seems to consider only the hypothesis of sudden annihilation, without mentioning anything like disappearance by elanguescence. However, a closer scrutiny of both the Phädon and the German Metaphysics reveals a further borrowing from the latter by the former.

In the Phädon, the first occurrence of the hypothesis of gradual extinction is strikingly different from its subsequent formulations. Mendelssohn introduces this alternative as one of the two possible ways in which we might conceive that the soul dies:

Either all its powers and faculties, its actions and passions suddenly cease, and the soul disappears as it were in an instant; or like the body, it undergoes gradual transformations [Verwandelungen], innumerable changes of appearance [Umkleidungen], which proceed in a continuous series, and in this series there is an epoch, where it has ceased to be a human soul, but has become something else, just as the body, after countless transformations, ceases to be a human body, and becomes dust, air, plant, or a part of another animal. Is there a third case in which the soul can die other than suddenly or gradually?

(95, translation modified)

After a few lines, Socrates reformulates the same alternatives by saying that "the soul may disappear suddenly, or gradually cease to be what it was" (95). Thus, the alternative that is subsequently described as a gradual loss or vanishing of the soul's inner power is initially presented as a gradual transformation. In this first presentation, to gradually die means to become something else, whereas in the [End Page 94] subsequent discussion it means to turn into nothing. Current readings overlook this difference or offer no explanation as to why Mendelssohn's argument shifts from gradual transformation to gradual annihilation.⁴⁹ Is it a real change of subject? Or is transformation meant to be somehow equivalent to annihilation? A glance at Wolff's doctrine of simple beings sheds some light on the issue.

After establishing that simple beings can cease to exist only by annihilation, Wolff's German Metaphysics addresses the objection that "a simple being could cease by transformation, insofar as a being having a different essence would result from it" (§103). This hypothesis is similar to Mendelssohn's first formulation. Although Wolff does not expressly characterize transformation as gradual, his example taken from the natural world shows that he has a gradual process in mind. People find the transformational hypothesis convincing, explains Wolff, because such transformations are supposed to actually take place in composite beings, "as when a tree develops from a leaf" (§103). If composite beings cease to be what they are by evolving into something different, why cannot simple beings undergo a similar fate? It is the same soul-body analogy that we have just found in Mendelssohn; the soul might cease to be a human soul and "become something else, just as the body, after countless transformations, ceases to be a human body, and becomes dust, air, plant, or a part of another animal." Wolff's reply to this objection is based on his doctrine of essences and the PSR. If the transformation has its reason in the thing's essence, then it is a mere change of state of a thing that remains essentially the same and does not cease to exist. If it has no reason in the thing's essence, then this essence would have to change in order for the thing to undergo such a transformation. But essences are immutable (§42); hence things cannot be transformed into essentially different things. Wolff's conclusion is that the alleged transformation of a simple being into something else would in fact consist in two distinct, unrelated events: first, the sudden annihilation of the former being, and second, the sudden creation of a new being from nothing, which confirms that "a simple being cannot cease otherwise than by being annihilated" (§103).

Sketching the alternative of sudden or gradual disappearance, the German Metaphysics is likely to have inspired Mendelssohn's similar distinction. In this perspective, Mendelssohn's shift from gradual transformation to gradual annihilation is not a complete change of subject, for annihilation is but one half of the process of transformation, the other half being the creation of another entity. If so, Mendelssohn's refutation of gradual extinction appears reminiscent of Wolff's refutation of the transformational hypothesis. Wolff's point is that radical transformation (unlike mere alteration) necessarily requires annihilation, so that ceasing to be by transformation is no real alternative to annihilation.⁵⁰ But since annihilation necessarily happens in an instant, even disappearance by transformation cannot avoid the leap from being to nothingness. Similarly, Mendelssohn's "On Sentiments" argues that, if God were to transform a plant's [End Page 95] bud into an insect, such a miraculous metamorphosis would not consist in a genuine transformation of the same entity (as when a seed develops into a tree), but rather in the annihilation of the former entity (the bud) and the creation of a new one (the insect).⁵¹ Some reasoning along these lines may have persuaded Mendelssohn that even gradual extinction must involve a final, nongradual step of abrupt annihilation. Thus, his Wolffian background prepared and oriented his attempt to use Boscovich's argument against discontinuous change to rule out not only sudden annihilation but also gradual extinction.

9. conclusion

The apparent intricacy of the Phädon's proof of the indestructibility of the soul reflects Mendelssohn's effort to intertwine multiple lines of reasoning stemming from different, and to some extent mutually independent, sources. Drawing on both Leibniz's view that death is not the real termination of life and Wolff's argument against the possibility of natural annihilation, Mendelssohn transferred Boscovich's doctrine of continuity from mathematics and physics to rational psychology. In spite of the Leibnizian emphasis on the continuity between life and the afterlife, however, the Phädon tends to depict the immortal soul as a pure spirit entirely separated from the body, in keeping with the Platonic inspiration of the dialogue. That is why it was crucial, for Mendelssohn, to specify what happens to the soul at the moment of the body's complete destruction, not only at the moment of its death. Although his deepest belief appears to have been that "no limited spirit can be entirely without a body,"⁵² in the Phädon he wittingly omits this esoteric Leibnizian doctrine and favors instead the more familiar view that the human soul is not forever united to an organic body.⁵³ In this way, Mendelssohn committed himself to the spiritualistic tenets of Wolffian rational psychology,⁵⁴ which were to become one of Kant's main targets in the first Critique.

It would be interesting to know whether Kant was in a position to recognize the Boscovichian model behind Mendelssohn's argument.⁵⁵ Although Kant never refers to Boscovich and possibly never read his works, it is not implausible that he [End Page 96] was acquainted with Mendelssohn's 1759 review of the Theory of Natural Philosophy.⁵⁶ This conjecture might help explain the content of his refutation of Mendelssohn's argument. Reviewing the first part of the Theory, Mendelssohn presents the Law of Continuity as the foundation of Boscovich's system (see JubA 5.1:58), then summarizes both the inductive and the metaphysical proof of the law. A series of changes that develops in continuous time must itself be continuous, for if a leap were to take place, then "the duration of the previous state would be really separated from the duration of the following state, and both boundaries, or the last instant of the previous time and the first instant of the following time, would be the closest to each other, which is absurd" (5.1:60, my translation). Although Mendelssohn omits Boscovich's discussion of annihilation, it is not impossible for the reader to recognize here the roots of the Phädon's argument against annihilation. Indeed, Kant's reconstruction of this argument might be reminiscent of the passage just quoted: if the soul suddenly ceased to exist, "there would be no time at all between a moment in which it is and another moment in which it is not, which is impossible" (Critique of Pure Reason, B 414).

One might further speculate that Kant's reading of Mendelssohn's review had informed Kant that in Boscovich's dynamics, forces transition from positive to negative values through their gradual reduction to zero (see esp. JubA 5.1:62). From this dynamic theory, Kant—unaware of Boscovich's distinction between zero and nothingness, not mentioned in the review—might have drawn inspiration for the idea of the gradual remission of the soul's powers that he uses to refute the Phädon's proof. In this way, he would have seized the opportunity to turn Mendelssohn's source against Mendelssohn himself. Of course, in the absence of textual evidence this must remain mere speculation. What appears more than conjectural is that there is no settling the issue of Kant's relation to Boscovich without also considering Mendelssohn's role in disseminating Boscovichian doctrines among the German public through his Phädon and his review of the Theory of Natural Philosophy.⁵⁷ [End Page 97]