This article is an introduction to Herbart's monadology. It discusses the fundamental concepts of his monadology and its similarity to Leibniz's monadology. A final section discusses the vexed question of Herbart's realism. It is argued that Herbart is more a transcendental idealist than a realist.
In the history of post-Kantian monadology, a prominent place should be given to Johann Friedrich Herbart (1776–1841). Along with Fichte, Schelling and Hegel, who were his contemporaries, Herbart was one of the great metaphysicians of the early nineteenth century. His metaphysics was a monadology in the most basic sense: it postulated the existence of a multitude of simple animate entities to explain the world. Although Herbart did not use the term ‘monadology’ to describe his metaphysics – the term had too many troubling associations by the early nineteenth century – and although he had some major disagreements with Leibniz, he still expressly saw his metaphysics as a rehabilitation of Leibnizian doctrine. His Allgemeine Metaphysik begins with the telling lines: ‘If there were any point in building a splendid door for the entrance to this work, we would find the materials for it from no one better than Leibniz’ (§1; VII, 21).1 As we shall soon see, however, Leibniz provided more than the building materials for the entrance to Herbart's metaphysics; he also supplied its foundation stones.
A contemporary reader might ask why he or she should bother with Herbart's monadology. It seems to be a hopeless attempt to revive antiquated doctrine, one thoroughly discredited by Kant's critique of metaphysics. But if we adopt this dismissive attitude, we only beg the question against Herbart. For he was well aware of the challenge that Kant had posed to metaphysics, and he was very concerned to meet it. Herbart even described himself as ‘ein Kantianer von 1828’,2 and he insisted that ‘any new metaphysics that comes forward as a science alone’ has to meet Kantian standards. Hence he understood metaphysics as ‘die Wissenschaft der Begreiflichkeit der Erfahrung’, as ‘ars experientiam recte intelligendi’. The task of metaphysics, in other words, was to determine the conditions of the possibility of experience. The first volume of Herbart's Allgemeine Metaphysik is a long discourse about what went right and what went wrong with not only pre-Kantian but also post-Kantian metaphysics; it also contains a diagnosis of the failure of the critical revolution itself. Herbart's central argument is that metaphysics can survive in the post-Kantian age only if it reinvents itself as a critical monadology. If we are to be fair to Herbart – and if we are to give a fair hearing to monadology itself – we do well to ponder that argument.
Apart from his monadology, any student of metaphysics has strong historical and philosophical reasons to take an interest in Herbart. For, though it has been nearly forgotten, Herbart's metaphysics was the contemporary of, and competitor to, that of the German idealists (Fichte, Schelling and Hegel) and that of the early romantics (Friedrich Schlegel, Novalis, Hölderlin and Schleiermacher). While the idealists and romantics found their inspiration in Plato and Spinoza, Herbart found his in Leibniz and Kant. Their metaphysics was monistic, teleological and idealistic (in the Platonic sense); Herbart's metaphysics was pluralistic, mechanistic and nominalist. A stronger antithesis is hardly imaginable. So, if we are interested at all in the philosophical basis of idealism and romanticism, we would do well to consider Herbart's criticism of these traditions. We understand and appreciate any philosophical tradition only when we know its antithesis.
Herbart's important historical place in the early nineteenth century raises the question: How could he be so forgotten? There are countless studies of the German idealists, singly and collectively; and there is a growing amount of work on the philosophy of the early romantics. But it is difficult to find today a solid and serious monograph on Herbart.3 Why? It is tempting to pin some of the blame for this on Herbart's difficult exposition, which places enormous demands on the reader's patience and understanding. Herbart said that he wrote for students already initiated into his philosophy, an audience which has long disappeared. But if troubling expositions alone were the reason for his neglect, the idealists and romantics would be just as forgotten. The reason for the ignorance of Herbart lies rather in an entrenched tradition of philosophical historiography. For generations, that tradition has stood under the shadow of Hegel's influential Geschichte der Philosophie,4 which portrayed the history of philosophy after Kant as the story of the development of his own system. Hegel simply wrote much of his opposition out of that story, according his opponents either a half-page (in the case of the romantics) or simply leaving them out entirely (Fries, Herbart and Beneke). Hegel's history is a wonderful story about his own intellectual development; but we cannot take it seriously as a general history of philosophy. Yet the major nineteenth-century historians after Hegel – Karl Rosenkranz, Johann Erdmann and Kuno Fischer – were Hegelians and therefore happy to accept their master's story as gospel (see Rosenkranz, Geschichte der kant'schen Philosophie; Erdmann, Die Entwicklung der deutschen Spekulation seit Kant; Fischer, Geschichte der neueren Philosophie, vols 5–7). They were followed in the twentieth century by Richard Kroner and Frederick Copleston (Kroner, Von Kant bis Hegel; Copleston, Fichte to Hegel) who, though no Hegelians, were content to repeat the historiography before them. Recent histories of nineteenth-century philosophy have, by and large, simply followed the Hegelian tradition (see e.g. Solomon, Continental Philosophy since 1750; Sandkühler, Handbuch Deutscher Idealismus; Bréhier, The Nineteenth Century).
Our task in this article will be to make up for some of the historical deficit in our understanding of Herbart. We will attempt to outline the basis of his monadology and show how it arose from his dispute with earlier traditions. Finally, we will examine one of the most misleading and persistent myths about Herbart's philosophy: its alleged realism. It should be obvious that all we can do here will be introductory and rudimentary. A proper understanding and appreciation of Herbart is the task for a future generation.
Herbart developed his metaphysics in self-conscious reaction against three opposing traditions: rationalism, idealism and romanticism. The rationalist tradition meant for him chiefly Leibniz, Descartes, Spinoza and Wolff, what we would today call pre-Kantian metaphysics. The idealist tradition, which was in formation in Herbart's day, was for him chiefly the work of Kant, Fichte and the young Schelling; Hegel, whom Herbart saw as a Schellingian, received little of his attention.5 The romantic tradition was typified for Herbart mainly by Schelling and Schleiermacher; he never mentions other romantics (viz. Friedrich Schlegel, Novalis, Hölderlin). To understand Herbart's metaphysics, we do well to consider his diagnosis of the failures of these traditions.
Herbart knew the idealist and romantic traditions well, for he had been for a long time in his youth an idealist; and he was, for a shorter time, even a romantic. His philosophical apprenticeship was under the idealism of Fichte. In his early Jena years (1794–97) he was regarded as Fichte's most talented disciple; but, gradually and painfully, he worked his way out of his teacher's legacy. The Fichtean ego, he realized by 1800, is only a fantasy (see his ‘Kritik der Ichvorstellung’, which was written in May 1800, in Sämtliche Werke I, 113–14). Already in 1806 he published the first version of his own metaphysics, Hauptpuncte der Metaphysik,6 which lays down the basis for much of his later system. It took another two decades, however, for his system to mature. The chief exposition of that system is his Allgemeine Metaphysik, which first appeared in two volumes from 1828 to 1829 (reproduced in volumes 7 and 8 of Sämtliche Werke).
Herbart's final settling of accounts with the idealist and romantic traditions appears in the first volume of his Allgemeine Metaphysik, which is a critical survey of metaphysics from Leibniz to Schelling. Herbart makes three fundamental criticisms of the idealists, two of which also apply to the romantics.7 First, a faulty methodology. The idealists, specifically Fichte and the young Schelling, inadvisedly followed Reinhold's method of beginning philosophy with a single self-evident first principle. It is absurd, Herbart argued, to limit the starting point of a philosophy down to a single proposition. We should begin enquiry from many different angles, and we should not limit ourselves down to one alone. No single proposition, in any case, is so fecund and replete with consequences that we can construct an entire system from it alone; we need at least several first principles to obtain more than trivial results. Even a set of first principles, however, is at best a means for organizing and systematizing our knowledge, Herbart insists, so that we should not confuse the order of exposition or knowledge, the ratio cognoscendi, with the order of being or things, the ratio essendi. Reinhold, Fichte and Schelling are guilty of just this fallacy, however, because they think that their first principle has not only epistemic but also ontological priority; they make the ego not only the first principle of knowledge but also the first principle of things (§98, VII, 174; §137, VII, 271). Second, confusion of the normative and factual. Because of their Reinholdian methodology, Fichte and Schelling sought for the unity of reason, a single principle to unite theoretical and practical reason. But, Herbart insisted, there is no such point of unity. The practical and the theoretical, the normative and the factual, are distinct forms of discourse which should not be confused with one another. No ‘ought’ ever follows from an ‘is’, no obligation ever flows from a fact. Because he sharply distinguished the normative from the factual, Herbart rejected teleology, which he regarded as a confusion of these forms of discourse. For similar reasons he questioned the organic concept of nature, which had been such an inspiration for Schelling and the young romantics. To assume that nature is an organism, he argued, is to presuppose that it has purposes or ends, that there is something that it should be. However, the best maxim of natural explanation is, as far as possible, to account for the mechanism of things. Teleological explanation is a mere stopgap until we have understood that mechanism (§160; VII, 324). Third, intellectual intuition. Seeing that the understanding or discursive intellect is limited to analysing concepts, and admitting that sensory intuition is restricted to sense experience, Fichte, Schelling and the romantics appealed to intellectual intuition as a source of knowledge. Allegedly, such an intuition gives us insight into an intelligible realm above or behind the realm of sense experience. Through it, we know either the absolute or the pure activity of the ego. But, for Herbart, the recourse to intellectual intuition is a desperate stratagem. It is difficult to persuade or convince anyone of the content of such an intuition if he or she does not have such a marvellous and magical faculty (§94; VII, 162). It is indeed almost impossible to communicate such intuitions; for to articulate them would be to describe them with concepts, which is to destroy their immediacy and simplicity. If we are to discuss and dispute, Herbart believed, there is no substitute for clear and distinct concepts. Philosophy is for him, as it was indeed for Kant, a discipline of conceptual thinking; hence he described it as ‘Bearbeitung der Begriffe’, i.e. the working through or analysis of concepts.8
Herbart's own metaphysics also derived from his diagnosis of the failures of pre-Kantian metaphysics. The chief fallacy of the old metaphysics, Herbart explains, was its conflation of essence with existence, i.e. its assumption that existence is only another property of a thing. The metaphysics of the Leibnizian–Wolffian school held that the actual or real is what is completely determinate (§5; VII, 23). The movement from the possible to the actual was then understood as one from the indeterminate to the determinate, as if to get to the existence of a thing one only had to add another property to it (§7; VII, 24–5). This conflation of essence and existence is also evident in Descartes and Spinoza, who, in one form or another, argue that God's existence follows of necessity from his essence. ‘Quo plus realitatis, aut esse unaquœque res habet, eo plura attributa ipsi competunt’ – those lines from Proposition IX, Part I, of Spinoza's Ethica were for Herbart the perfect epitome of the chief fallacy of rationalist metaphysics (Spinoza, Opera·Werke II, 96. ‘The more reality or being a thing has, the more attributes belong to it’). It was Kant's greatest contribution to the critique of metaphysics, Herbart claimed, to have shown the fallacy in this old way of thinking and to have sharply distinguished essence from existence (§32; VII, 55).
The net effect of this fallacy was that traditional metaphysics had become too removed from experience and existence. It assumed that its analysis of concepts was sufficient to understand reality, though there was no guarantee that these concepts had any reference at all.
A new metaphysics, Herbart was convinced, would have to avoid this fallacy by beginning with the given in experience and by analysing its most basic concepts. The fundamental task of metaphysics, as ‘the science of the conceivability of experience’, was to investigate the forms of experience, both metaphysically and psychologically (§93 Anm; VII, 159). It would ask a) how these forms arise (the psychological investigation) and b) what validity they should have (the metaphysical explanation).
There can be no doubt whatsoever, Herbart acknowledges, that Kant's philosophy marked a revolution in modern metaphysics. Kant not only saw through the fallacy of the old metaphysics – its conflation of essence and existence – but he also rightly stressed the need for metaphysics to remain within the limits of experience (§70; VII, 119). The starting point of all future metaphysics, Herbart believes, is Kant's thesis that existence is not a property but simply the absolute positing of a thing. It was another great contribution of Kant, Herbart adds, when he distinguished metaphysics from ethics, theoretical from practical reason, ‘is’ from ‘ought’ (§39 Anm.1; VII, 77). But, as important as these criticisms were, Kant still blunted their full force by advancing other doctrines that fed suspicious metaphysical speculation. His doctrine of the unity of reason blurred the distinction between theoretical and practical reason; and the concept of organism of the third Kritik not only confused the normative with the factual but it also encouraged speculation about nature as a vast living being (§39 Anm.1; VII, 77). Chiefly because of the third Kritik, Herbart claims, Kant thwarted his own reform of metaphysics.
Although Herbart thinks that Kant's philosophy was a great step forward, he also regrets that Kant did not come on the scene earlier when the old Leibnizian–Wolffian metaphysics was less decrepit (§99; VII, 176). That would have been enough to reform that metaphysics without destroying it. It could then have been built on the more solid foundation of existence and experience. Reclaiming that old metaphysics, and basing it on such a critical foundation, was Herbart's chief philosophical mission.
What was that old metaphysics? The monadology, of course. In the first chapter of the first volume of his Allgemeine Metaphysik Herbart sees the history of metaphysics since Kant as chiefly an attempt to reform Leibniz's monadology (§25 Anm; VII, 39). It is a far-fetched view of post-Kantian philosophy, perhaps, but it at least shows the importance that Herbart gave to Leibniz's doctrine. For him, the need to return to the monadology was clear and pressing: ‘There is no need for a power of divination, not even for much true knowledge of metaphysics, to see that the monads, once their opponents die off, must come back to life once again’ (§116; VII, 218).
What, specifically, did Herbart see in Leibniz's monadology? First and foremost, it was Leibniz's analysis of reality into simple self-sufficient units, entities which could serve as the ultimate elements or constituents of things. Like Leibniz, Herbart believed that there had to be such units; otherwise, analysis would proceed ad infinitum and there would be no reason to regard anything as real. Again following Leibniz, Herbart insisted that we could not understand these units in terms of extension, mass or impenetrability (Anm. VII, 43). These physical dimensions arose from the composition or aggregation of these units, so that they could not be attributed to any unit alone or in its intrinsic nature. On these grounds, Herbart argued for something like a Leibnizian distinction between reality and appearance: reality consists in simple real beings, and appearance consists in their composition or aggregation, which is the work of the perceiver. Again no less than Leibniz, Herbart understood his units as animate beings, as having a power of life (§374; VIII, 284–5). Hence he attributes to all real things, all ultimate units of reality, what he calls ‘self-preservation’ (Selbsterhaltung), that is, the power to resist other things intruding on them (§244; VIII, 117–18). This does not mean, however, that all monads have a power of representation. While all representation presupposes self-preservation, not all self-preservation takes the form of representation (see Herbart, Lehrbuch zur Einleitung in die Philosophie, §131; IV, 225). Because he saw self-preservation as the fundamental force of every being, Herbart criticized Kant for assuming that repulsion is a basic force behind matter. In postulating this force Kant took impenetrability as a datum, though it too stands in need of explanation (§154; VII, 299). Although Kant was right in criticizing the old view of matter as dead extension (§150; VII, 293), he did not go far enough, and he should have questioned the fact of impenetrability. Behind that impenetrability, and behind that force of repulsion, there lies another deeper power: self-preservation, the drive of everything to preserve or maintain itself, to repel forces impacting upon it.
So Herbart has great debts to Leibniz's monadology. But he also makes two criticisms of it, objections so basic that it is necessary to distinguish his monadology from that of his great predecessor. First, Herbart does not accept Leibniz's teleology, more specifically his assumption that a monad is a nisus or entelechy, a force that strives for self-realization (§132; VII, 262). While Herbart, in attributing self-preservation to real things, endows them with a primitive kind of life, he still wants that life to be understood on a mechanical basis. Second, Herbart maintains that there can, and indeed must be, interactions between monads, so that they stand in causal relations with one another; in other words, his monads have ‘windows’.9 While the intrinsic nature of a monad remains forever the same, impervious to all change, its extrinsic nature or properties depend on its relations to other things (§214; VIII, 76). Herbart thinks that Leibniz was perfectly correct to reject the doctrine of interaction as it was understood in his own day: an ‘influx’ where one body somehow gives its properties to another, as if properties could be somehow free-floating and detach themselves from one thing and attach themselves to another (§116; VII, 218). But Herbart maintains that we can have interaction without influx, provided that we make the right distinction. Namely, we must distinguish between real and apparent causality, where real causality is independent of time and place, and where apparent causality provides only a rule for succession for events in time (§146; VII, 287). That we must attribute real causality to substances Herbart leaves no doubt; he makes it a basic maxim ‘No substantiality without causality!’ (§220; VIII, 83).10 Herbart had a strong motivation for attributing interaction to monads: it was only on that basis that he could explain the latest findings of chemistry. He is very much impressed with the advances of this new science, and he insists that metaphysics must come to terms with them (§378; VIII, 287) Accordingly, the final section of his Allgemeine Metaphysik is devoted to the explanation of chemical affinities on the basis of his monadology. To explain them, Herbart attributes to his monads relations of attraction and repulsion or what he calls ‘mutual interpenetration’. Clearly, we are far removed from Leibniz's self-enclosed and self-absorbed monads.
Now that we have some idea of what metaphysics should not be according to Herbart, we should proceed to examine what it should be. Let us attempt to reconstruct, in its barest outlines and from the ground up, Herbart's monadology. The foundational part of Herbart's metaphysics is what he calls its ‘ontology’, which appears in the first section of the second volume of his Allgemeine Metaphysik. We will not provide a commentary on this section – the task for an entire volume – but only pull together those ideas central to its foundation.
Herbart's ontology begins with some reflections on the starting point and method of metaphysics. It was the mistake of classical metaphysics, we have seen, to have begun with the concept of possibility, and to have neglected the connection of its concepts with reality (§166; VIII, 15–16). To rectify this mistake, to avoid constructing castles in the air, metaphysics must begin, Herbart demands, with reality itself, i.e. with the facts of experience, with what is given in sense perception (§161; VIII, 11). The proper method of metaphysics, just like that of natural science, is both analytic and synthetic (§164; VIII, 13). It must first analyse its facts, determining their basic components and their relations; it must then synthesize these components, putting them together again in their proper relations. It will then confirm its analysis if its synthesis reconstitutes the facts from which it began. Proper method in metaphysics is shaped like an arc, Herbart says. It begins with the given; it descends to the depths to explain the reality behind the given; and finally it returns back to the given (§164; VIII, 14). We shall hold Herbart's metaphysics to the demands of his own methodology; we will raise the question: Can it really explain the facts of experience?
Because metaphysics must cling to reality, Herbart insists that it has to begin with an analysis of the concept of reality itself (§199; VIII, 44). We must determine, in other words, what it means to ascribe reality to something. We are forced to do this, Herbart explains, because the sceptic gives us good reasons to doubt that the given – what appears to us in sensation – is real. Appearance and reality are the two poles of metaphysics (§193; VIII, 47). If what appeared were reality pure and simple, there would be no need at all for metaphysics. If we are to find reality, we must go beyond and get behind appearances; yet we still have to begin with appearances. As Herbart summarizes the predicament of the metaphysician: ‘We doubt the reality of the given; we seek being; and our only hope of finding it still depends on the given’ (§198; VIII, 52).
What, then, do we mean by reality? We have already seen that reality or existence does not add anything to the content of a concept (§202; VIII, 56). To say that something exists is not to attribute a new property to it. What exists first appears to us in experience as a thing with its properties. What we sense are the properties of something, for example, I sense that this apple is red, round, sweet and hard; but the thing itself, what unifies these properties, is not given in experience (§201; VIII, 56). These sensed properties presuppose some reality – the thing itself – of which they are the appearances (§199; VIII, 53). Reality then seems to be something in which properties inhere.
When we say that this thing is real, Herbart maintains, we assume that it has an independent existence. We assume that its existence does not depend on the existence of any other thing (§204, VIII, 59–61; §244, VIII, 177). If it were to depend on the existence of something else, it would not be entirely or completely real by itself, but it would be real only conditionally or hypothetically. If something is to be unconditionally or absolutely real, we must ‘absolutely posit’ its existence, so that its existence is ‘full and complete’ (§72; VII, 122). Because each fully real and complete being has such an independent existence, it is entirely indifferent to the existence of other things; it could just as well exist without these things; their existence is completely contingent for it (§244; VIII, 177). The absolute positing of a thing concerns its substance, what exists prior to its properties or accidents (§73; VII, 125).
This substance or real being, Herbart further insists, is absolutely simple (§73, VII, 124–5; §207, VIII, 63). It is one thing, completely integral and indivisible. For Herbart, like Leibniz, the very concept of existence involves that of unity. Leibniz made it an axiom: ‘That what is not truly one being is not truly one being either’ (Leibniz to Arnauld, April 1687, in Die philosophischen Schriften, II, 97). With that axiom Herbart fully agrees. It follows from it, he thinks, that we cannot take away anything from, or add anything to this unity, because addition and subtraction involve composition or decomposition, the assumption that ultimate reality has parts (§73; VII, 124).
At this point, we do well to ask ourselves: why does Herbart assume that there is a plurality of real beings? He points out that the concept of being itself is neutral about number, and that there is no reason from it alone to postulate one or many things (§208; VIII, 66). Yet the existence of many things is a fundamental premise of any monadology, and it is a premise that Herbart shares with Leibniz. Why? Part of the answer lies in Herbart's starting point: the given of ordinary experience. What is given to us in experience, he notes, are diverse complexes of characteristics, where each complex appears as one thing (§165; VIII, 15). These complexes are not free-floating or arbitrarily interchangeable because they appear together in our experience independent of our will and imagination (§169, VIII, 19–20; §171, VIII, 21). They therefore appear to inhere in something which unites them and makes them just one thing. It seems, then, that we should posit the existence of one substance or one thing for each complex. This is part of Herbart's reasoning, which is explicit in this passage from his Einleitung in die Philosophie:
From the truly one, said Lucretius, there will never be many; from the truly many there will never be one. But many is given; therefore one must presuppose an original manyness.
(§119; IV, 194)
But this answer only goes so far, because we could demand to know why these apparently many things are not just modes of a single universal substance. What makes us attribute independent reality to them? Herbart's answer to this question is apparent from his reaction against Spinoza's monism, which appears in the first volume of the Allgemeine Metaphysik. There Herbart makes it clear that there can be no transition from Spinoza's single substance to the modes of ordinary experience (§§46–48, VII, 51; §89, VII, 97). There is an unbridgeable gulf between the unity and eternity of substance on the one hand, and the multiplicity and temporality of its modes on the other hand. We cannot have, Herbart argued, both a one and a many: if the one becomes many, it divides itself and ceases to be one (§110; VII, 200). And so, if we are to explain appearances, the apparent reality of a multitude of things, we have no recourse but to embrace pluralism. We can do so, furthermore, with no need to worry about limiting the reality of substance. Spinoza feared that if there were two substances, then one would lose its reality by being limited by another. But this is a groundless worry, Herbart argues, because no substance loses any of its reality because of another (§73; VII, 125).
So far, Herbart has come as far as Leibniz in the first paragraphs of his Monadologie. He has postulated the existence of a plurality of simple substances, each of which is independent in essence and existence. All other reality is made of complexes of these simple substances. From the same basis as Leibniz, Herbart then draws very similar conclusions: that his simple substances are non-spatial and eternal. They are non-spatial because space is a form of composition, which comes from adding real things together (§265; VIII, 155). They are also eternal because time involves change, which means adding to or subtracting something from the reality of a thing (§281; VIII, 172). Space and time are therefore forms of the appearance of things, Herbart argues, because their reality depends on composition and quantity. In general, Herbart argues that we cannot attribute quantity to real beings because that would imply that they are divisible (§78, VII, 122;§208, VIII, 66). All quantities, all forms of aggregation or composition, concern only the relations between things but not the intrinsic reality of things themselves.
What about qualities? Do they not also involve some kind of division of the ultimate units of reality? We call our apple red, round, sweet and firm, where all these qualities are separable from one another. Indeed, they are so distinct that it seems impossible to refer to a single thing that unites them. Herbart fully recognizes this point, and insists that his ultimate units are an indivisible unity. When we see a thing as having different properties, he maintains, that is the work of our perceiving consciousness, which makes these distinctions to make sense of its experience (§213; VIII, 75). The utter unity and simplicity of substances does not mean, however, that they are bare substrata, completely bereft of any qualitative dimension. They have qualities; it is just that they are all bound together in an indistinguishable unity or whole.11
In accounting for the inherence of properties in things, Herbart develops one of his most distinctive and peculiar concepts: ‘accidental views’ (zufällige Ansichten). An accidental view of a thing arises from the perceiving consciousness, which attributes properties to the thing from its various perspective upon it. Herbart likens these accidental views to the solutions of a mathematical problem (§190; VIII, 45). They are similar to the various numbers which, when added together, give a certain sum; given these numbers, the sum is necessary, the only possible result. There are, however, many different numbers that could result in the same sum; and hence the view is ‘accidental’, because it is not the only set of numbers that gives the sum. There are as many accidental views of a substance as there are sets of numbers that could give a definite sum.
Ultimately, Herbart faces the same fundamental problem as Leibniz: How do we construct matter, space and time from the reality of simple things? The problem is pressing because these simple things are spaceless and timeless, so that adding them together, forming complexes from them, should not result in space and time. Herbart's solution to this problem is similar to Leibniz's: he attributes the reality of space and time to the perceiver. The forms of composition arise from the consciousness of the subject, which joins together the multitude of independent things so that they appear in space and time (§265; VIII, 154). Herbart insists that neither space nor time, neither motion nor change, are properties of things themselves; they are only appearances of things for some consciousness. They belong to what Herbart calls the sphere of ‘objective appearance’, where an objective appearance consists in how any intelligent being must perceive a plurality of independent real things (§293; VIII, 187). An objective appearance differs from a subjective one in that it must hold always and for all perceivers, whereas a subjective appearance holds only sometimes for some perceivers. What Herbart means by objective appearance is pretty much what Leibniz calls a ‘phaenomena bene fundata’, i.e. an appearance of a thing based on its reality, though it is not a picture or perfect image of a thing.
To give some idea of the difficulty Herbart faces in reconstructing appearances from real beings, it is instructive to consider his construction of matter in Part II of his Allgemeine Metaphysik. To explain matter we have to account for cohesion, the joining of simple entities to form a whole. For two entities to come together, they must interact. But to assume interaction requires that things somehow penetrate one another (§271; VIII, 161). There must be, as Herbart puts it, ‘imperfect togetherness’ or ‘partial penetration’ of one thing by another. There must be penetration because if things were entirely impenetrable, they would not attach to one another; nothing would aggregate. There must also be partial penetration, because if they were entirely penetrable then one thing would become coincident with the other; they would be congruent and one and the same thing (§267; VIII, 158–9). But, though it is indispensable, Herbart admits that the concept of partial interpenetration is a fiction because it has to assume that simple entities are somehow divisible, that one part of them is penetrated but not the other. Real beings, however, do not have parts (§278; VIII, 167). Herbart then asks himself in a telling passage: Why assume imperfect togetherness, partial penetration, if it is only a fiction? His answer is disarming: we must assume it because real things form matter, and the existence of matter is a phenomenon that we must explain (§278; VIII, 167). The construction of matter therefore requires admitting a fiction, indeed a contradiction (i.e. that the indivisible is divisible), into our ontology.
The same problem arises for Herbart's construction of ideal space. Herbart distinguishes between ideal and real space: ideal space is that which arises from the interaction of monads themselves; real space is that which appears in our ordinary sense experience, and which arises from the aggregation of monads by our sense organs and consciousness (see Lehrbuch zur Einleitung in die Philosophie §134; IV, 235). Herbart's construction of ideal space, which appears in Chapters 2–4 of the third section of his Allgemeine Metaphysik, is one of the most complicated and subtle theories of the entire book. After the most exhausting and exacting account of how to form lines, planes and solids from the interaction of monads, Herbart finally arrives at a three-dimensional space akin to that of ordinary experience. Herbart is confident that, if we have the proper psychological explanation, we can ultimately identify this ideal space with the real space of our ordinary experience (§264; VIII, 152–3). Yet the derivation, for all its subtlety and sophistication, is a failure. Herbart's real space, like cohesion, presupposes the partial interpenetration of monads, and so it involves the same difficulties as that concept (§265; VIII, 155).
So, in the end, Herbart has as much difficulty in accounting for the realm of appearance – space, time and motion – as Spinoza. His metaphysics did not have the arc shape he demanded: it descended into the depths of the ultimately real; but never again did it ascend to the realm of appearance. The only way it could lift itself back to that level was by resorting to fictions, by dividing the indivisible and by connecting the unconnectible.
In his popular study of Herbart, Otto Flügel, a noted Herbart scholar, once wrote: ‘He who knows only a little about Herbart knows that he was a realist at a time when his age mostly thought idealistically’ (Herbarts Lehren und Leben, 1). ‘Realism’ has indeed been the catchword to describe Herbart's philosophy. The signal doctrine of his philosophy, according to the textbooks, is its ‘realism’, which is supposed to be its distinguishing feature from all the idealist systems of his day (see e.g. Ueberweg, Grundriss der Geschichte der Philosophie, 107–27). There must be something to the interpretation, one would think, given that Herbart himself would sometimes refer to his philosophy as ‘my realism’ (Herbart to Brandis, 1 Oktober 1831, Sämtliche Werke VIII, 412).
Any attentive reader of the above paragraphs will be rightly puzzled by this interpretation, however. How, he or she will ask, can Herbart be a realist if he denies the reality of matter as an extended, solid thing? How, indeed, can he be described as such if he maintains that the world of space and time are only an appearance of the ultimately real for the conscious subject? Is not such a doctrine better described as a form of idealism?
There is indeed a problem here. The realist interpretation of Herbart's philosophy stands in need of drastic qualification. To give the reader a sense for the problem, consider the following lines from Herbart's Einleitung in die Philosophie:
We are completely enclosed inside our concepts; and for just that reason, because we are so, our concepts decide the real nature of things. Whoever holds this to be idealism (from which it is completely different) must know that, according to his usage, there is no other system than idealism.
(§114; IV 183)
Though Herbart himself does not adopt this usage, he still affirms the doctrine behind it, a doctrine that would usually be regarded as idealism. Why not, then, just call his philosophy ‘idealism’? There is all the more reason for doing so, when we consider that there are other passages from Herbart's works where he endorses doctrines that can be described only as ‘idealism’. Thus he maintains explicitly, as Kant once had, that the immediate objects of our knowledge are only representations, that we construct an external world from them, and that we cannot get outside our representations to know things-in-themselves (see Lehrbuch zur Einleitung in die Philosophie, §103; IV, 159–60; and Hauptpuncte der Metaphysik §3, II, 192). Herbart thinks that these are fundamental truths of idealism, which no philosophy can or should bring into question.
All this poses the question: What, if anything, does ‘idealism’ and ‘realism’ mean in Herbart's philosophy? In what sense is his philosophy realist, and in what sense is it idealist?
As much as Herbart affirms a kind of idealism, he also sometimes distances himself from it, and he even talks about the need for ‘a refutation of idealism’. It is just such passages that seem to lend fuel to the realist interpretation of his philosophy. What exactly Herbart meant by ‘idealism’ becomes clear from section four of his Allgemeine Metaphysik, which is devoted expressly to exposing the deficiencies of idealism. It becomes clear from Herbart's exposition that ‘idealism’ has a very specific meaning. It refers to the idealism of Fichte's Wissenschaftslehre, which Herbart, following Jacobi, regarded as the purest and most radical version of idealism (see Allgemeine Metaphysik, §141; VII, 276. See also Lehrbuch zur Einleitung in die Philosophie, §104; IV, 162. See the ‘Anmerkung’ to the second, third and fourth editions). Fichte's idealism represents the boldest form of idealism because it sees everything as the product of the self-positing ego. Herbart argues, however, that such an idealism is implausible for the simple reason that the purely self-positing ego cannot explain the reality of the non-ego (Allgemeine Metaphysik, §§323–6, VIII, 228–33). The ego is held to be perfectly self-positing; but it also posits a non-ego, which is opposed to itself. How does that which is purely self-positing limit itself and posit something opposed to itself? How do we derive from a purely self-positing ego something self-oppositing? We cannot, Herbart complains, and not least for this reason we should abandon pure idealism.
But this is only the beginning of the problems for idealism, according to Herbart. He further argues that not only the non-ego in general, but also the variety of qualities given in our experience – the so-called manifold of experience – cannot be derived from the ego. What sense qualities appear to us, and when and how they appear, is strictly contingent for the purely formal activities of the absolute ego. Herbart stresses that it is not only the matter of experience that is given, i.e. the simple existence and quality of sensations, but also its form, i.e. in what order these sensations appear to us (Allgemeine Metaphysik, §118, VII, 223 and §201, VIII, 55). The specific properties that appear to our senses are simply given to us, and they occur in just this order rather than another, completely independent of our will and imagination. When we see our apple as red, round, hard and sweet, for example, we cannot substitute these properties with any others; precisely what we see, and where, when and how we see it, depends on factors beyond conscious control. What we see is partly the result of things-in-themselves, which exist independent of our consciousness, and partly the result of perceptive organs and activities, so that what we see are appearances of things-in-themselves.
It should be clear from these arguments that we cannot describe Herbart's philosophy as an unqualified idealism. They show that Herbart thinks that idealism, in its pure Fichtean version, is an utterly unworkable and implausible philosophy at odds with the facts of experience. The facts of experience are the great stumbling block of idealism, because they show that not only the matter but also the form of experience is given to us.
But it should also be clear that Herbart's philosophy also cannot be described as an unqualified realism. Herbart is not a direct realist, i.e. someone who holds that we have an immediate knowledge of the external world, because he does not think that our representations directly reflect, copy or mirror given characteristics; all that we are directly aware of, he insists, are our own representations. But Herbart is also not an indirect realist, that is, someone who thinks that we can infer the characteristics of things-in-themselves from our representations. There are two problems with this position: first, that the nature of our representations depends so much on the psychological and physiological apparatus with which we perceive things that it becomes impossible to separate out what does and does not represent reality itself; and, second, that reality in itself is very different from the reality that we perceive – it is eternal and spaceless, whereas what we perceive is in space and time. Finally, Herbart is also not a scientific realist, that is, someone who assumes that the formal, quantifiable or mathematical properties of things are their real properties; for, as we have seen, he denies that reality in itself is quantifiable and that it exists in space and time. For Herbart, the quantifiable aspects of a thing are indeed its least real aspects.
There is still, however, an important sense in which Herbart's philosophy can be described as realist after all: Herbart thinks that there is a realm of things-in-themselves, and that these things-in-themselves exist apart from and prior to our consciousness or perception of them. Although we cannot ascribe to things-in-themselves, as they exist apart from and prior to our experience, the sense properties that we know from experience, we are still justified in regarding these properties as appearances of things-in-themselves, that is, as how they appear to human beings with their sensibility and forms of understanding. These appearances are, in part, the effects of how things-in-themselves act upon us. In this rather minimal sense Herbart is no more a realist than Kant, who also expressly maintains the reality of things-in-themselves and who protests the conflation of his philosophy with idealism on just these grounds. Still, we can understand Herbart's insistence on the term ‘realism’ when we remember Kant's own difficulty in distinguishing his transcendental idealism from Berkeley's idealism. In a revealing passage from his Hauptpuncte der Metaphysik Herbart claims that Kant made a mistake in calling his doctrines of the nullity of space, time and motion, and ignorance of things-in-themselves, a form of idealism (§10; II, 205). These doctrines are a form of realism, in his view, because they still presuppose the existence of things-in-themselves.
If we strictly define Kant's transcendental idealism in its intended sense – as a the doctrine of the distinction between appearances and things-in-themselves – then Herbart's own philosophy is very close to Kant's. In a remarkable passage from his Allgemeine Metaphysik Herbart virtually admits as much, stating that his doctrine can be described as ‘idealist’ and that it is closer to Kant's views more than anyone else (§298; VII, 193). In one respect, though, Herbart departs from Kant in providing a more unequivocal basis for his realism: he thinks that we are perfectly justified in extending the categories beyond sense experience; he denies that the principle of causality has only a temporal meaning, as if it held only for succession in time (see Lehrbuch zur Einleitung in die Philosophie, §115; II, 184. See ‘Anmerkung 2’ to the second, third and fourth editions. Also see Allgemeine Metaphysik, §76, VII 126; §146 and VII 287; and §299, VIII 193). Hence it is possible to infer the existence of things-in-themselves and that they are the causes of our sensations.
The heart of Herbart's realism and idealism appears in his concept of ‘objective appearance’, which he expounds in his Allgemeine Metaphysik (§§202–3; VIII, 186–8). The realm of objective appearance appears in space and time, where space and time are appearances because they depend on the perceiving subject, but where they are also objective because they are universal and necessary characteristics of these appearances. Although Herbart claims that his doctrine is akin to Kant's idealism more than anyone else, it would have been more accurate for him to have said that it is closer to Leibniz. For he holds, just as Leibniz once had, that the appearances of things in space and time are due to the effects of things-in-themselves on our sensibility, and that space and time are confused representations of things-in-themselves (see Einleitung in die Philosophie, §95; IV, 147. See the ‘Anmerkung’ to the second, third and fourth editions). Appearances are indeed for Herbart, just as they were for Leibniz, necessarily tied to things-in-themselves, so that they are attached to them as appearances of things-in-themselves; they cannot be detached from them as if they were nothing more than representations in consciousness. What we perceive in our experience is ‘an accidental view’ of the thing-in-itself, i.e. it is an aspect or property of that thing as it appears to our perceiving consciousness. It is indeed telling that Herbart objected to Kant's own distinction between appearances and things-in-themselves on the grounds that it left a too sharp separation between them (Allgemeine Metaphysik, §299; VIII, 195). Here Herbart targets Kant's frequent references to appearances as mere representations in us.
All in all, then, it is a rather slender and minimal sense of ‘realism’ that Herbart advocates. It amounts chiefly to two claims: (1) that there exist things-in-themselves apart from and prior to consciousness; and (2) that experience consists in appearances of them. Still, we can understand why Herbart would want to call this doctrine realism. It was an important distinguishing characteristic of his philosophy in contrast to Fichte's, which was seen as the most consistent and radical form of idealism in his day. Herbart, who had once been a Fichtean idealist, was very keen that his philosophy not be confused with the idealism of his former mentor. What better way to keep a distance from him than by calling his philosophy realism? Yet, for all the reasons we have seen above, the epithet become very misleading, disguising Herbart's own fundamental idealist doctrines.
It is significant that Herbart had classified Leibniz as a realist (Allgemeine Metaphysik, §115; 211). He was a realist for the same reasons as Herbart. He too affirmed the reality of things-in-themselves; he too claimed that we perceive appearances of them; and he too claimed that they are the basis for our perception of an external world. All the more reason, then, that Herbart would want to identify with Leibniz. This was one final affinity of his philosophy with that of the great rationalist, whose doctrines he had done so much to rehabilitate.
Bréhier, E. The Nineteenth Century: Period of Systems 1800–1850. Translated by Wade Baskin, 111–203. Vol. 6 of The History of Philosophy. Chicago: University of Chicago Press, 1968.
Cassirer, E. Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeit. Die nachkantische Systeme. Berlin: Cassirer, III, 378–410.
Copleston, F. Fichte to Hegel. Vol. 7 of A History of Philosophy. New York: Image Books, 1965.
Erdmann, J. Die Entwicklung der deutschen Spekulation seit Kant. Vols 5–6 of Versuch einer wissenschaftlichen Darstellung der Geschichte der neuern Philosophie. Stuttgart: Frommann-Holzboog, 1977.
Fischer, K. Geschichte der neueren Philosophie. Vols. 5–7. Heidelberg: Winter, 1872–77.
Flügel, O. Herbarts Lehren und Leben. Zweite Auflage. Leipzig: B.G. Teubner, 1912.
Hegel, G. W. F. Werke. Edited by Karl Michel and Eva Moldenhauer. Frankfurt: Suhrkamp, 1971.
Herbart, J. F. Hauptpuncte der Metaphysik. Göttingen: J.C. Baier, 1806.
Herbart, J. F. Sämtliche Werke. Edited by Karl Kehrbach and Otto Flügel. 19 vols. Langensalza: Hermann Beyer & Söhne, 1887–1912. Reprinted: Aalen, Scientia Verlag, 1989.
Kant, I. Schriften. Edited by Prussian Academy of Sciences. Berlin: de Gruyter, 1902.
Kroner, R. Von Kant bis Hegel. Tübingen: Mohr, 1921.
Leibniz, G. W. Die philosophischen Schriften von Gottfried Wilhelm Leibniz. Edited by C. J. Gerhardt. Berlin: Weidmann, 1875–90.
Röd, W. Geschichte der Philosophie. Munich: Beck, 2006.
Rosenkranz, K. Geschichte der kant'schen Philosophie. Leipzig: Voss, 1840.
Sandkühler, H. J., ed. Handbuch Deutscher Idealismus. Stuttgart: Metzler, 2005.
Schmitz, J. N. Herbart Bibliographie 1842–1963. Winheim: Julius Beltz, 1964.
Solomon, R. C. Continental Philosophy since 1750: The Rise and Fall of the Self. Oxford: Oxford University Press, 1988.
Spinoza, B. Opera·Werke. Darmstadt: Wissenschaftlicher Buchgesellschaft, 2008.
Ueberweg, F. Grundriss der Geschichte der Philosophie. Vierter Theil. Das Neunzehnte Jahrhundert, Neunte Auflage, edited by Max Heinze. Berlin: Mittler und Sohn, 1902.
1 All references to Herbart will be to Sämtliche Werke. References to this edition will be to volume (Roman numeral) and page number (Arabic numeral). References to the Allgemeine Metaphysik will be first to the paragraph number (§) and then the volume and page number. ‘Anm.’ designates ‘Anmerkung’, a remark appended to a paragraph. This work consists in two volumes, volumes 7 and 8 of Sämtliche Werke. Volume 7 consists in paragraphs §§1–160 and volume 8 in paragraphs §§161–444. All translations from the German are my own.
2 See the ‘Vorrede’ to Allgemeine Metaphysik, VII, 13. See also Herbart's academic lecture Oratio ad capessendam in academia georgia augusta professionem philosophiae ordinariam habita, SW X, 53–64, where he states (63): ‘Kantianum ipse me professus sum, atque etiam nunc profiteor … ’.
3 A computer search through the WorldCat for all publications on Herbart from 1950 until 2014 yielded 1462 items, though almost all of them were new printings or editions of his works or studies of his theory of education. Only a handful, which for reasons of space we cannot mention here, were on his philosophy proper. For a bibliography of works on Herbart before this period, which were more numerous, see Schmitz, Herbart Bibliographie. The best treatment of Herbart's metaphysics is still that of Cassirer, Das Erkenntnisproblem in der Philosophie, III, 378–410.
4 Hegel, Vorlesungen über die Geschichte der Philosophie, volumes 18–20 of Werke. The third section of volume 20, 314–462, ‘Neueste deutsche Philosophie’, treats what we would now call the post-Kantian systems.
5 Herbart knew Hegel well enough. He made some passing remarks about his philosophy in the preface to the second volume of his Allgemeine Metaphysik (VIII, 6). He wrote a highly critical review of Hegel's Philosophie des Rechts for the Leipziger Literatur-Zeitung in 1822 (Nr. 45–47), Sämtliche Werke XII, 140–54. Herbart saw Hegel as part of the Schellingian school and saw that as the chief problem with his philosophy.
6 A second edition appeared in 1808 in Göttingen with Justus Friedrich Danckwerts. Both editions are reproduced in Sämtliche Werke, II, 175–226.
7 The romantics (Friedrich Schlegel, Novalis and Hölderlin) did not accept the idealist methodology of first principles; they were in their early years severe critics of Reinhold's Elementarphilosophie. Herbart's first criticism of the idealists’ methodology would therefore not apply to them. The romantics did, however, accept teleology and intellectual intuition, so the second and third criticisms would hold also for them.
8 See Herbart's definition of philosophy in Lehrbuch zur Einleitung in die Philosophie, Sämtliche Werke IV, 38–9. The reference to work or ‘Arbeit’ in ‘Bearbeitung’ is deliberate. Famously, Kant had spoken of the need for ‘Arbeit’ in philosophy, which he understood as the labour involved in the analysis of concepts. He contrasted such work with a more lazy philosophy that would appeal to feelings and intuitions. See his ‘Von einem neuerdings erhobenen vornehmen Ton in der Philosophie’. Herbart was siding with Kant against Schelling and the romantics.
9 In his admirable summary of Herbart's philosophy, Röd maintains that Herbart's monads have no windows. See volume IX/1 of Geschichte der Philosophie, 188. However, Röd does not note the important distinction between real and apparent causality. He is also forced to recognize (188) that Herbart does need to postulate some form of interaction between his monads.
10 Herbart's emphasis. This maxim seems difficult to understand in view of Herbart's critique of causal interaction in his Lehrbuch zur Einleitung in die Philosophie. See §106; IV, 164–6. But Herbart's arguments against the reality of interaction here should be understood as a critique of the influx theory.
11 This is the point Herbart would make against Cassirer's criticism, Das Erkenntnisproblem, III, 401, that his real beings are propertlyless and therefore ‘the caput mortuum of abstraction’. His reals have characteristics or properties; it is just that they are dissolved in a unity, and that we have to grasp them separately and therefore not as they are in themselves.