Jeremy Bentham, Social Criticism & Levels of Meaning^*

Dr Colin Tyler
University of Hull

Contents
Introduction
§1 The Metaphysical Foundations of Censorial Utilitarianism
§2 Two Levels of Meaning
§3 Problems Caused by Bentham's Theory of Language
Conclusion

'In every part . of the common field, concomitant and correspondent to Eudaemonics considered as an art, runs Ontology, considered as a science.’

Jeremy Bentham, 'Essay on Nomenclature and Classification'¹

Introduction

For many years, scholars underestimated the significance of Jeremy Bentham’s claim that his metaphysics was the foundation of his utilitarianism. For example, Elie Halévy’s classic study makes no substantial reference to Bentham’s metaphysics.² Thirty-four years later, Mary Mack did describe Bentham’s writings on logic (of which his metaphysical - or as he called them ‘ontological’ - writings formed integral parts) as ‘Bentham’s most important work’, although in 1962 this was ‘virtually unknown’.³ Yet for years afterwards, none of the major studies of Bentham’s utilitarianism made substantive reference to his metaphysical beliefs.⁴

The situation is now quite different with Ross Harrison, Douglas Long, Gerald Postema, Bernard Jackson, William Twining, Philip Schofield, and others arguing vehemently for the reliance of Bentham’s legal and political thought on his metaphysics.⁵ For example, Long claims that, ‘it would be difficult to overstate the influence exerted by Bentham’s metaphysical and epistemological beliefs and theories on his political ideas.’⁶ Similarly Philip Schofield argues that, ‘[i]t is only through appreciating Bentham’s originality in ... [his theory of real and fictitious entities] that historians will be able to construct a framework of interpretation which convincingly explains the emergence of Bentham’s political radicalism, and indeed of philosophic radicalism more generally.’⁷ Such a shift of emphasis has implications for every area of Bentham scholarship, but most particularly for the study of the critical branch of his legal and political thought, or, as it can be labelled, his ‘censorial utilitarianism’ (sc. social criticism with a Benthamite purpose). From this perspective a number of questions arise about his theory of social criticism, including:

(1) What philosophical function, if any, does Bentham’s metaphysics perform within his censorial utilitarianism?, and,

(2) Can a metaphysical foundation be found for Bentham’s censorial utilitarianism?

Two points should be noted about this third question. First, it is less demanding than question one, as it is answered once a fundamental flaw is found within Bentham’s metaphysics. Second, assuming the failure of the metaphysics actually left to us by Bentham, we are not then required to consider whether another metaphysical theory could ground his censorial utilitarianism, as we are with question two. There are other, more positive reasons for considering the third question, not least because it functions as either a prolegomenon or a partial answer to the first two. Moreover, it is an intrinsically important historical and philosophical question whether or not Bentham was successful in his attempt to lay solid foundations for his utilitarian ‘edifice of granite.’⁸

Focusing on Bentham's most concerted work on metaphysics, which he undertook between 1813 and 1816, this article argues that the manner in which Bentham develops and employs this distinction created very significant problems for his censorial utilitarianism. The argument has two stages. Section one outlines the metaphysical foundations that Bentham hopes would justify censorial utilitarianism. Section two argues that Bentham ascribes two levels of meaning to terms within propositions, and that this dualism creates insurmountable difficulties for his theory of social criticism. Section three throws further doubt on the coherence of these theories of meaning, by noting Bentham's apparently necessary reliance on semi-real entities, and arguing that such reliance radically undermines his distinction between real and fictitious entities. It is concluded that, as Bentham's theory of social criticism presupposes the viability of the first level of meaning, this collapse entails the collapse of the censorial utilitarianism which rests upon this mode of social criticism.

§1 The Metaphysical Foundations of Censorial Utilitarianism
return to contents

Douglas Long observes that for Bentham,

'To "perfect" knowledge is not merely to augment it quantitatively but also to refine it in quality. To augment knowledge was in itself, as Bentham saw it, the function of the "expositor." Refining knowledge was the special task of the "censor." The active process of advancing the cause of Enlightenment began with censure - the criticism of established ideas.'⁹

The phrase ‘established ideas’ needs to be given its broadest meaning here. The reformer’s censorial function is not reducible to simply the reasoned critique of the language and perspectives embodied within legal, political and social institutions. Consequently it is not essential that in practice every critical citizen is a good metaphysician. A censor can legitimately expend his energies on such prosaic matters as gauging the honesty or otherwise of public functionaries, the efficiency of particular departments of state or particular judges, or a multitude of matters of public policy.¹⁰ Yet Bentham is emphatic that such a non-metaphysical censor is constantly in danger of being led astray by common-place ‘prejudices and wild conceits’.¹¹ It is for this reason that the metaphysical stage of censorial utilitarianism must be carried out by some social critics. For example, he argues that that dangers of the 1791 French Declaration of the Rights of the Man and the Citizen spring from the fact that it is ‘a perpetual vein of nonsense flowing from a perpetual abuse of words.’¹² Similarly he writes in Scotch Reform (1808): ‘Fiction (in law) [is] a wilful falsehood, uttered by a judge, for the purpose of giving to injustice the colour of justice.’¹³ The censor should be aware moreover that language is not even at its most dangerous when false propositions masquerade as true ones. Towards the end of A Fragment of Government (1776), Bentham reflects on his analysis of the common law theory of the rights and duties of the supreme power: ‘Had the doctrine been but false, the task of exposing it would have been comparatively an easy one: but it was what is worse, unmeaning; and thence it came to require all these pains which I have been here bestowing on it’.¹⁴ In short, Bentham insists throughout his writings that the censor should always bear in mind that figurative, obscure and ultimately meaningless language tends to blind citizens to their true interests, thereby leaving them open to manipulation by those in power. Consequently the censorial function relies on - even if it cannot be reduced to - the ‘demystification’ every institution and practice which wears the mask of authority so as to disguise the operation of sinister interests.¹⁵ Metaphysical criticism by some censors must constantly be used to bring others back to that more fundamental knowledge which comes only through the careful exercise of individual reason on the language of public action and justification.¹⁶

It is for this reason that, for example, Bentham concludes his 1814 essay on logic with the discussion which appears in the Bowring edition of his Works as (the vast bulk of) ‘Logical Arrangements.’¹⁷ Bentham notes the utilitarian efficacy of making reference to political fictitious entities such as ‘obligation’ and ‘right’, but emphasises that:

Of either the word obligation or the word right, if regarded as flowing from any other source [than pleasure or pain], the sound is mere sound, without import or notion by which real existence in any shape is attributed to the things thus signified, or no better than an effusion of ipse dixitism.¹⁸

Bentham understands language and power as being inextricably linked, then. To counter the potential abuse of power, the censor's task is to analyse and, where necessary, to reform the established or officially sanctioned relationships between the constituent elements of language. The censor should clear the 'mist' that masks the logic - and ultimately the meaning - of propositions, a logic which would otherwise surreptitiously privilege certain practical conclusions.

Bentham argues that the key to censorial utilitarianism can be found in his analysis of the relations between entities which we believe exist independently of our belief in them (which he called 'real entities') on the one hand, and entities which we believe exist only as terms denominating imagined particular objects within our discourses (which he called 'fictitious entities') on the other. More formally, Bentham defines a ‘real entity’ as ‘an entity to which, on the occasion and for the purpose of discourse, existence is really meant to be ascribed.’¹⁹ He defines a ‘fictitious entity’, on the other hand, as ‘an entity to which, though by the grammatical form of the discourse employed in speaking of it existence is ascribed, yet in truth and reality existence is not meant to be ascribed.’²⁰

Notice that Bentham rests the distinction between these two types of entities not on their respective ontological statuses, but on the agent’s beliefs about their respective ontological statuses. Following from this important fact and adopting Dummett’s categories, I have argued elsewhere that Bentham adopts an anti-realist metaphysics, rather than the realist metaphysics that is usually imputed to him.²¹ Dummett states his classic distinction between realism and anti-realism in the following terms.

'Realism I characterise as the belief that statements of the disputed class possess an objective truth-value, independently of our means of knowing it: they are true or false in virtue of a reality existing independently of us. The anti-realist opposes to this the view that statements of the disputed class are to be understood only by reference to the sort of thing which count as evidence for a statement of that class.'²²

Much of what follows will develop this reading and pursue its implications for Bentham’s censorial utilitarianism. Before doing so however, it important to note a further distinction made by Bentham in 1814-1816 manuscripts.

One must be careful about stating the relationship of the concept of ‘a fictitious entity’ to another widely noted concept that is employed by Bentham - 'a fiction'. First, most commentators have in effect agreed with Mary Mack’s judgement that, ‘In Bentham’s vocabulary, all words that are not concrete or proper names are fictions.’²³ This view is mistaken, at least in regard to the 1814-1816 manuscripts (which have formed the basis for many discussions of his theory of fictions). In fact, Bentham saw fictions as heuristic devices which enable the use of fictitious entities in thought and discourse; fictions are ways of conceiving imaginary relations between objects rather than the objects themselves. More formally, Bentham defines a ‘fiction’ as ‘the mode of representation by which the fictitious entity thus created, in so far as fictitious entities can be created, are dressed up in the garb, and placed upon the level, of real ones.’²⁴ The second point to notice about Bentham's 'fictions' is that he did not wish individuals to dispense with their use in discourse: he merely wished people to be aware of their existence, ontological status and their proper role in language and thought, and the dangers attendant upon their use. Once again, this point has not always been fully appreciated in the secondary literature. For example Charles Everett has implied that Bentham regarded every legal fiction as ‘a deliberate or unconscious attempt to conceal reality, to mislead the reader.’²⁵ This is not Bentham's position. Even though the censor should recognise the dangers of such illusions, Bentham does not wish to see figurative language banished completely, in favour of a conceptual and linguistic system based on literal truth and reality.²⁶ As with most of Bentham’s philosophical positions, the justification for retaining some reference to fictitious entities and fictions is found in an appeal to the greatest happiness principle. Fictions and fictitious entities are required for any language and consequently any thought beyond that of ‘the brute creation’.²⁷ Judged by the standard of utility, discourse using propositions that only referred to real entities would be too cumbersome when compared to the circumspect use of some propositions that refer to fictitious entities. Without such propositions, there could be no widespread medicine, no art, no poetry, no engineering, no chemistry or other sciences, and while there could be none of the present linguistically based judicial corruption, similarly there could be no possibility of an uncorrupted system of law, rights and justice. In short, some legal fictions, like some legal fictitious entities, may be justified on the ground that their careful use tends to bring a net utilitarian benefit.

So, Bentham argues that without a clear and solid understanding of real and fictitious entities as well as fictions, a censor could not take command of language, the facts expressed in language, and therefore could not guide the world in accordance with the greatest happiness principle. Crucially the censor would be unable to carry out the twin processes of ‘exposition’ (sc. analysis of propositions and terms used within propositions) and ‘methodization’ (sc. arrangement or classification), activities which ‘will, in every branch of art and science be found capable of affording useful direction and eminent assistance.’²⁸ Methodization in particular (which Bentham sees as almost the whole of logic, and certainly its core)²⁹ renders ‘indisputably valuable assistance’ to all forms of study and activity: ‘In relation to art and science without distinction, teaching, learning, and improving - in relation to art, practising.’³⁰ He even argues a little earlier in the 1814 essay on logic that, ‘it is the province of Logic to take command and give direction to the course of Ethics itself.’³¹ In short, Bentham understands a successful theory of logic to be a necessary precondition of censorial utilitarianism, and a rational metaphysics as a precondition of successful exposition, methodization, and logic more generally.³²

Consequently, Bentham is required provide at least the outlines of a coherent method with which to distinguish between and then analyse particular fictitious entities, fictions, and fallacies, as well as a method with which to translate particular propositions employing fictitious entities into propositions making reference solely to real entities. Bentham presents the relevant metaphysical foundations in his 1814-16 ‘Ontology’ and ‘Logic’ essays, and outlines the analytical procedure in his important ‘Essay on Nomenclature and Classification’ (written in 1815, published in 1817).³³

It is these texts that I will focus on in what follows. As I have critiqued Bentham's main expository tools (paraphrasis, phraseoplerosis and archetypation) elsewhere, and so will not examine them here.³⁴ Instead, I will argue in the following two sections that, unfortunately, there are other reasons for believing that Bentham's analysis of real entities, fictitious entities and fictions is fatally flawed.

§2 Two Levels of Meaning
return to contents

It has been noted above that Bentham believes fictitious entities derive their meaning from their relationships to real entities. Consider the following example.

An obligation (viz. the obligation of conducting himself in a certain manner,) is incumbent on a man, (i.e. is spoken of as incumbent on a man) in so far as, in the event of his failing to conduct himself in that manner, pain, or loss of pleasure, is considered as about to be experienced by him³⁵

It is meaningful to speak of a person as being under an obligation (a fictitious entity) if and only if the consequence of his failure to act in the way prescribed by the obligation would cause him to suffer pain (a real entity). A person has a legal obligation if and only if his failure to act in the manner prescribed by law (or his performance of an act proscribed by law) would cause a representative or embodiment of the legal system (for example, a policeman, judge, bailiff, prison warder or hangman) to inflict some form of pain upon him.

I will refer to this as the 'first level' of meaning (fictitious entities derive their meanings from their relations to real entities). It is so well covered in the literature that further comment is unnecessary here. What has not been so frequently or clearly recognised is that, at times, Bentham invokes a second level of meaning. If the first level is characterised as 'vertical' - in that ultimately fictitious entities derive their meanings from their relationships to the more fundamental real entities - the second level can be seen as 'horizontal', in that the meaning of every meaningful term that is employable in discourse is a function of its place within a system of meaningful terms. Importantly, this second level of meaning obtains for entities as we conceive of them.

Bentham believed that truth and clarity are preconditions of achieving the greatest happiness. Yet ‘truth’ is a notion that he rarely seeks to analyse in his metaphysical or logical writings, although its realisation is central to his censorial utilitarianism. It is clear that Bentham believes truth necessarily entails coherence within and between propositions.³⁶ Yet, coherence is not a sufficient condition of truth. The proposition ‘A unicorn is eating the grass outside my window’ is perfectly coherent in itself. Nonetheless, there is no evidence that unicorns exist anywhere except as imaginary creatures in the minds of storytellers and the writers of articles. What can be said is that for Bentham a proposition is true only to the extent that it carries a meaning which accurately represents the matters of fact that it purports to represent. Bentham asserts that the notion of ‘a fact - or a matter of fact' presupposes the notion of 'existence'.³⁷ Yet he argues also that ‘the word existence is in all cases the name of a fictitious entity.’³⁸ In this same manuscript, he writes: ‘Take away all other qualities, this [quality of existence] remains: to speak more strictly, take any entity whatsoever, real or fictitious, abstract the attention from whatsoever other qualities may have been found belonging to it, this will still be left.’³⁹ It is unclear what this statement means given Bentham’s metaphysical presuppositions. At times, Bentham’s discussion of ‘existence’ is clearly confused: ‘Being as above a species of quality, existence itself is a fictitious entity: it is in every real entity: every real entity is in it.’⁴⁰ Here, Bentham has effectively abandoned the priority rule of meaning which he believes is required to make fictitious entities intelligible. No longer is it the case that fictitious entities derive their meaning ultimately from the reference which they bear to real entities (the first level of meaning). Existence (a fictitious entity) imparts meaning to ‘every real entity’. Indeed, Bentham’s difficulties with this concept of ‘existence’ are so great that at one point he is reduced to stating that ‘the punster who has played with [the concept] nothing till he is tired may renew the game with existence and non-existence.’⁴¹

The problem is that if Bentham rejects the priority rule of meaning by holding to this conception of existence, then his distinction between real entities and fictitious entities appears to collapse, thereby radically undermining the censor, the methodizer, and therefore censorial utilitarianism as a whole.

This strange tension is indicative of a very peculiar strand of Bentham’s metaphysical writings. It is possible to get a clearer understanding of this feature by turning to his well-known distinction between perceptible and inferential entities. Bentham claims in 1814 that, ‘Speaking of entities, ideas might perhaps be accordingly considered as the sole perceptible ones; substances, those of the corporeal class, being with reference and in contradistinction to them no other than inferential ones.’⁴² Immediately after this passage, he states: ‘But if substances themselves be the subject of the division, ... it is to corporeal substances that the characteristic and differential attributive perceptible can not but be applied, the term inferential being thereupon employed for the designation of incorporeal ones.’⁴³

> It now appears that the label of perceptible entity refers to a ‘differential attributive’, rather than an absolute one. Presumably the same categorisation applies to the label inferential entity. Both are, in other words, relative terms. But what factors determine which label is appropriate in any particular situation? ‘Speaking of entities’, corporeal substances are inferential, whereas ideas (sc. incorporeal substances) are perceptible. Yet, ‘if substances themselves be the subject of the division’, then corporeal substances are perceptible entities, while incorporeal are inferential.⁴⁴ Here, it is the perspective of the classifying agent which determines the perceptible-inferential status of the entities under consideration, rather than the agent’s ‘persuasion’ of the entity’s ontological status.

There are a number of arguments which Bentham could employ in support of this perspectival account. One possible approach would be to assert what I shall call ‘the doctrine of relative perspicuity’. Here, the appropriate categorisation of two or more objects is determined by the relative strength of the evidence marshalled in the subject’s consciousness in favour of the claims to existence of the respective entities. This reading is supported by Bentham’s illustration of this passage: corporeal entities tend to cause more pain than do incorporeal entities if you try to ignore their existence.⁴⁵ Consequently, ‘the inference [of the formers existence] ... is much stronger and more irresistible than the [latter] inference.’⁴⁶

Unfortunately, this move contradicts key elements of his earlier explication of the distinction between perceptible and inferential entities. The doctrine of relative perspicuity ignores any link between ontological status and perceptibility. Indeed, it contradicts the explication which he gives at the start of the sequence which is currently under discussion: ‘perceptible those [entities] of the existence of which the persuasion is produced by sense without reasoning, i.e. without reflection: inferential those of the existence of which the persuasion, in as far as it has place, is produced by reason, by reflection.’⁴⁷ When employing the doctrine of relative perspicuity, an indisputably fictitious entity, such as quantity, would be a perceptible entity in comparison to what is for Bentham an indisputably fabulous entity such as a unicorn or the Devil. Moreover, where impressions would be perceptible entities, corporeal substances would be merely inferential entities in comparison. Strong impressions would be perceptible when compared with weaker impressions, which in turn would be inferential. Remember that according to the doctrine of relative perspicuity, it is the relative length of the chain of reasoning which determines the most appropriate categorisation of entities. Yet, in the other two cases, it is the presence or absence of any chain of reasoning. In the former case, the labels are relational; in both of the latter cases, they are absolute.

Let us stay with the doctrine of relative perspicuity for the moment. It is significant that an ambiguity that is analogous to that noted above arises in relation to Bentham’s analysis of real and fictitious entities. With regard to ‘Place’ Bentham observes that, ‘Considered as a modification of Space, it would, like that stand upon the footing of the name of a real entity. Considered as a species of relation, it would stand upon the footing of a fictitious entity.’⁴⁸ It appears now that the censor’s perspective determines both the object’s perceptible-inferential status and its real-fictitious status. In both cases, the perspectival account contradicts Bentham’s more formal characterisation of the respective bifurcate divisions.

The perspectival account prioritises the agent’s subjective attitudes to entities in drawing the respective distinctions. Nevertheless, the perspectival account is logically incompatible with an anti-realist reading. Where the perspectival account has been shown to rest on a comparison of two or more entities, an anti-realist reading involves no such necessary comparisons, relying as it does solely upon the agent’s subjective judgement of the ontological status of the objects of consciousness. The former is a relative distinction whereas the latter is absolute. The simultaneous operation of these incompatible methods within Bentham’s metaphysics tends to render unintelligible the real-fictitious entity distinction. These problems are compounded when one turns to Bentham's theory of language.

§3 Problems Caused by Bentham's Theory of Language
return to contents

Many commentators have believed Bentham’s most significant innovation in linguistics to be his argument that meaning is a property of propositions rather than of isolated terms. W V Quine observed that, for Bentham ‘words are seen as imbibing their meaning through their use in sentences.’⁴⁹ Quine judged this ‘shift from terms to sentences’ to be one of the ‘five milestones of empiricism’.⁵⁰ He noted Bentham’s anticipation of Bertrand Russell’s theory of contextual definition, a technique which Bentham ‘applied to contextual not just to grammatical particles and the like, but even to some genuine terms, categorematic ones.'⁵¹ Quine even claimed that Bentham’s recognition of ‘the semantic primacy of sentences’⁵² inaugurated ‘a revolution in semantics’ on the scale of Copernicus.⁵³ Bernard Jackson has noted that Bentham ‘anticipated Russell, Vaihinger, the Vienna Circle and the Wittgenstein of the Tractatus.’⁵⁴ Herbert Hart argued much the same thing (although without Quine's clichéd invocation of Copernicus) when he highlighted Bentham’s claim ‘that sentences not words are the unit of meaning’ and that Bentham's 'doctrine of Logical Fictions’⁵⁵ anticipated ‘the ideas of Logical Constructions, Incomplete Symbols, and Definition in Use which are a marked feature of Bertrand Russell’s philosophy and the forms of analytical philosophy which stem from it.’⁵⁶ On this basis, Hart concluded that Bentham’s logical writings ‘are [not] only of value as so many blows against [political] reaction and revolution. There are indeed many things of great speculative importance in them.’⁵⁷

It is important to bear in mind that Quine and Hart were concentrating on Bentham’s contribution to linguistics. The relationship of his linguistics to his distinction between real and fictitious entities is complex, but vital to our present concerns. It is indisputable that, for Bentham, on at least one important level all entities, whether real or fictitious, are objects (physical or psychical) which are denominated by noun substantives: they are the objects denominated, not the terms which denominate those objects.⁵⁸ In this way entities per se are not necessarily creations of language. Clearly, if censors are to analyse, criticise and reform them, then entities must be denominated, given that analysis is in part but necessarily a linguistic operation. Nevertheless, there is nothing in Bentham’s conception of an entity per se which requires its members to be denominated.

This can be said to obtain at the level of the genus.⁵⁹ When one moves to the level of species, the situation becomes more complex. Imagine that I am persuaded of the purely corporeal existence of this rock. I am then logically required to regard the rock's existence as a non-linguistic fact about the world. The rock is a real physical entity. Similarly, imagine that I am persuaded that I perceive a feeling of pleasure. Once again, the existence of that impression does not presuppose of necessity its existence as a linguistic fact (where I name it and am able to refer to it in discourse). The feeling of pleasure is a real psychical entity. Alternatively, where I believe that an entity’s existence does necessarily presuppose language, logically I am required also to affirm that the entity is in part - but again necessarily - the creation of language: it is, in short, fictitious.

We can now turn to the philosophical relationships between fictions and fictitious entities, relationships which, as we will see, have very significant implications for Bentham's censorial utilitarianism. The semantic primacy of sentences logically commits him to the claim that the essence of each fictitious entity necessarily presupposes the import that it bears as part of meaningful propositions. Almost all propositions of this type presuppose 'a mode of representation' (sc. a fiction) that constitutes a key facet of what would now be called the hermeneutic context in which fictitious entities are employed. In other words, the construction of meaningful propositions presupposes a framework of interrelated ideas from which meanings can be drawn (when needed) via particular communicative acts that make reference to propositions, which in turn make reference to fictitious entities. This implies that, as a linguistic manifestation of thought, every fictitious entity denominated by a noun substantive of necessity presupposes relations to the imports which it bears within a system of possible meaningful propositions.

The role of relations in this theory is pivotal. ‘No two entities of any kind can present themselves to the mind at the same time’, writes Bentham, ‘- no, nor can so much as the same object present itself at different times, without presenting the idea of relation.’⁶⁰ Bentham concerns himself in the sequence of manuscripts from which this passage comes with the ‘most extensive and in [their] conception simple of all relations’: place (including space and body), time, motion, rest, action, passion, subalternation ‘viz. logical subalternation’, opposition, causation, and existence with its several modifications or correspondent fictitious entities - non-existence, futurity, actuality, potentiality, necessity, possibility and impossibility’.⁶¹

The notable point is Bentham's contention that as soon as two or more entities, whether real or fictitious, are perceived, conceived or remembered simultaneously, the mind produces a relation between them. Even if these entities are conceived as being radically different from one another, nevertheless they must be understood by their relationships to one another. For example Bentham argues that, ‘as identity is but the negation of diversity, thence, if on no occasion diversity had ever been, neither on any occasion would any such idea as that of [identity] come into existence.’⁶² The relational nature of conceived entities obtains even where entities are seen as being in some sense identical, as for example where the objects are ‘this book now’ and ‘this book as I perceived it yesterday’.

This general claim that all conceived entities (whether real or fictitious) are inherently relational is significant given that Bentham goes on to state that as soon as it appears in the mind, ‘the fictitious entity called Relation, ... swallow[s] up all the others. Every other fictitious entity becomes but a mode of this one.’⁶³ Now, it is unclear precisely what it is for ‘Every other fictitious entity’ to be merely ‘a mode’ of Relation. Bentham appears to be claiming that it is part of the essence of each particular member of a set⁶⁴ of conceived entities that the agent perceives at a given time (call that set CE1) to be understood as necessarily related to the other members of CE1. Certainly, there is nothing to prevent CE1 changing: the mind can merely modify the group of entities which form it, for example by altering their number or even their interrelationships. Nevertheless, such changes necessarily alter the conceived essence (the meaning) of each particular member of the set, because the ideas by which each one is constituted are then understood to stand in different relationships to the other ideas within that set. Consequently, the meaning of every particular conceived entity in some sense and to some degree necessarily presupposes its relations to the other entities that are simultaneously present in the agent’s consciousness.

It is in this sense that Bentham defends a second level of meaning. The first level pertains most obviously to the meaning of fictitious entities: the meaning of the latter derives ultimately from their relationships to real entities. The second level of meaning concerns not only fictitious entities, but real entities as well. This level is so inclusive due to Bentham's recognition of the fact that the act of conceiving an object (apprehending it, or remember it, or imagining it, and so on) requires the subject to relate that object to the other objects of consciousness. Crucially, these relations form part of the meaning which the object has for the subject.

If it is possible to maintain Bentham's distinction between real and fictitious entities, then the preceding contention does not obtain for real entities: even though the mind can only conceive of such objects in relation to other entities, that does not entail that the agent believes the real essence of the real entity is even partially constituted by those relationships. This picture would be completed by noting that fictitious entities would necessarily presuppose real entities, but real entities would not necessarily presuppose fictitious entities.

Yet there are several problems in maintaining that distinction. Here I will focus on Bentham's discussion of the concept of ‘space’. Analysing the notion of ‘this or that individual portion of Space’ is problematic for Bentham in that, as with existence, the concept seems to possess both real and fictitious facets. On the one hand, space is a particular immediate object. On the other hand, of necessity it possesses ‘limits’ as well as form, quantity, and ‘even motion’, all of which are fictitious entities. For example he claims that, ‘Substances being real physical entities, perceptions real psychical entities, matter, form, quantity, and so on, so many fictitious entities, both descriptions being in part applicable to space, neither of them applicable entirely, space may be regarded and spoken of as a semi-real entity.’⁶⁵ At one point Bentham even goes so far as to treat space as both an ‘individual portion’ and 'in the aggregate.’⁶⁶ Some commentators have glossed over this serious difficulty, however doing so ignores a significant ambiguity in the real-fictitious entity distinction.⁶⁷

A second difficulty is exemplified in Bentham observation in his essay on what he refers to as 'ontology' that, ‘There not being any real entity to represent, the entity represented [by the name ‘motion’] can not be any other than fictitious.’⁶⁸ If there is no ‘real entity to represent’, then strictly speaking ‘motion’ is a fabulous entity, like a unicorn, which creates a new incoherence within Bentham’s metaphysics. On the one hand, Bentham is committed to the view that motion, along with those other concepts which he believes presuppose it such as cause and effect, are necessarily required in human discourse. Yet, if motion is a fabulous entity, then it is a member of 'the other class of unreal entities' (that is, other than the class of fictitious entities). As such it can ‘be spared’ in meaningful discourse beyond the level of ‘the brute creation’.⁶⁹ The real-fictitious distinction seems once again to have proved deficient as a mode of logical analysis.

Conclusion
return to contents

The most significant implication of these various flaws is that the essence of each entity is determined by the particular and essentially contingent set of relations to other entities in which that entity stands in the particular mind of the particular agent. Consequently, once again there is no absolute standard, set of concepts or ‘reality’ that Bentham can use to ground either his metaphysics or his censorial utilitarianism.

Endnotes

^{* I wish to thank Professors Hugo Bedau, Philip Schofield and William Twining, as well as two anonymous referees, for their many helpful comments on earlier drafts of this paper. The responsibility for what follows is, of course, mine alone. I adopt the standard system of referencing the Bentham archive at University College London: ‘UC’ = refers to University College archive, roman numerals refer to the box, and arabic numerals refer to the folio (for example, UC ci.015).}

^{1 'Chrestomathia: Appendix No. IV Essay on Nomenclature and Classification', (hereafter, ENC), Bowring viii, pp. 63-123, reprinted in J Bentham, Chrestomathia, edited by M J Smith and W H Burston (Oxford: Clarendon, 1993) (hereafter, Chrestomathia), pp. 139-276, p. 181}

^{2 Elie Halévy, The Growth of Philosophic Radicalism, translated by M Morris (London: Faber and Faber, 1928).}

^{3 Mary P Mack, Jeremy Bentham: An Odyssey of Ideas 1748-1792 (London: Heinemann, 1962), p.162.}

^{4 Ross Harrison, Bentham (London: Routledge & Kegan Paul, 1983), chapter 3, David Lyons, In the Interests of the Governed (Oxford: Clarendon, 1973; revised edition 1991), Frederick Rosen, Jeremy Bentham and Representative Government (Oxford: Clarendon, 1983), Lea Campos Boralevi, Bentham and the Oppressed (Berlin: de Gruyter, 1984), Paul J Kelly, Utilitarianism and Distributive Justice (Oxford: Clarendon, 1990), and Janet Semple, Bentham’s Prison (Oxford: Clarendon, 1993). Other notable studies which understate the significance of this area of Bentham’s thought include David Baumgardt, Bentham and the Ethics of Today (Princeton, NJ: Princeton, 1952), Nancy Rosenblum, Bentham’s Theory of the Modern State (Cambridge, Mass: Harvard, 1978), and L J Hume, Bentham and Bureaucracy (Cambridge: CUP, 1981) although see pp.190-1, 291n94.}

^{5 Douglas G Long, Bentham On Liberty (Toronto: University of Toronto Press, 1977). Gerald Postema, ‘Facts, Fictions, and Law: Bentham on the Foundations of Evidence’ in W Twining, ed., Facts in Law (Wiesbaden: Franz Steiner Verlag Clarendon, 1986). Bernard Jackson, ‘Bentham, Truth and the Semiotics of Law’, in M Freeman, ed., Current Legal Problems 1998, Volume 51: Legal Theory at the End of the Millenium (Oxford: Clarendon, 1998). William Twining, Theories of Evidence (London: Weidenfeld and Nicholson, 1985). Philip Schofield, ‘Political and Religious Radicalism in the Thought of Jeremy Bentham’, History of Political Thought, vol. XX, no. 2 (1999).}

^{6 Long, 1977, p.210, see also ibid., pp.206-220 passim.}

^{7 Schofield, 1999, p.291.}

^{8 Jeremy Bentham, Rationale of Judicial Evidence, in The Works of Jeremy Bentham, John Bowring, general editor, 11 vols. (Edinburgh, 1838-1843), (hereafter, Bowring), vii (hereafter, RJE), p.599. On the historical significance of Bentham’s ontological and linguistic theories, see Harrison, 1983, chapter 3, and Emmanuelle de Champs, 'The Place of Jeremy Bentham's Theory of Fictions in Eighteenth Century Linguistic Thought', Journal of Bentham Studies, vol. 2 (1999).}

^{9 Long, 1977, p. 14.}

^{10 Rosen, 1983, chapter 2, passim.}

^{11 RJE, p.599.}

^{12 Jeremy Bentham, ‘Nonsense Upon Stilts, or Pandora’s Box Opened, … ’, in J Bentham, Rights, Representation, and Reform: Nonsense upon Stilts and other writings on the French Revolution, edited by P Schofield, C Pease-Watkin and C Blamires (Oxford: Clarendon, 2002), p.321 (UC cxlvi. 065). See Long, 1977, pp.61-62, 66-69.}

^{13 Jeremy Bentham, Scotch Reform, in Bowring, v, p.13.}

^{14 Jeremy Bentham, A Fragment on Government, chapter V, paragraph 13.}

^{15 Herbert L.A. Hart, Essays on Bentham (Oxford: Clarendon, 1982), pp.21-39.}

^{16 RJE, p. 599.}

^{17 Jeremy Bentham, ‘Logical Arrangements, or Instruments of Invention and Discovery Employed by Jeremy Bentham’, Bowring, vol. III, pp. 286-295. The material which the editors took out of the context of Bentham’s 1814-16 logic essay is printed at ibid., pp. 286-295. The manuscripts for ‘Logical Arrangements’ are at British Library Additional Manuscripts 33,550, fos.2-27, although the editors added fos. 2 and 3 to the final chapter of the 1814-16 ‘Logic’ essay.}

^{18 'Logical Arrangements', Bowring, iii, p. 293.}

^{19 Jeremy Bentham, De L’ontologie et autres texte sur les fictions, P Schofield, J-P Cléro and C Laval, eds. (Éditons du Seuil, 1997) (hereafter, Ontology), p.164. UC cii. 016. 7 July 1821.}

^{20 Ontology, p.164. UC cii. 016. 7 July 1821.}

^{21 Colin Tyler, '"A Foundation of Chaff"? A critique of Bentham's metaphysics, 1813-186', British Journal of the History of Philosophy (forthcoming in 2005), §1.}

^{22 Michael Dummett, 'Realism (1963)' in his Truth and Other Enigmas (London: Duckworth, 1978), p.146.}

^{23 Mack, 1962, p.1.}

^{24 Ontology, p.84. UC cii. 023. 23 September 1814.}

^{25 Charles W Everett, Jeremy Bentham (London: Weidenfeld and Nicolson, 1969), p.60.}

^{26 See, for example, Harrison, 1983, chapters III and IV, Long, 1977, pp.128-129, Baumgardt, 1952.}

^{27 Ontology, p.87. UC cii.023. 23 September 1814.}

^{28 'Logic', pp. 261-262.}

^{29 'Logic', 261.}

^{30 'Logic', pp. 262, 261.}

^{31 'Logic', p. 222.}

^{32 This is one reason why, in Long’s words, ‘The perfection of knowledge was to be achieved by this synthesis of the censorial and expository functions.’ (Long, 1977, p. 13).}

^{33 ENC, pp. 254-273, especially pp. 257-258.}

^{34 Tyler, 'Chaff', §4.}

^{35 'Logic', p.247.}

^{36 For example: ‘whatsoever therefore were the considerations by which he was engaged to endeavour to persuade himself of the truth of the self-contradictory and therefore impossible proposition, remain without any thing to counteract their force.’ (Ontology, p.156. UC cii. 077. 3 October 1814.)}

^{37 UC cii. 301. 2 August 1814. Also see UC cii. 535. 14 December 1815.}

^{38 Ontology, p.150. UC cii. 074. 2 October 1814.}

^{39 Ontology, p.150. UC cii. 074. 2 October 1814.}

^{40 Ontology, p.152. UC cii. 074. 2 October 1814.}

^{41 Ontology, p.152. UC cii. 074. 2 October 1814.}

^{42 Ontology, p.180. UC cii. 015. 25 September 1814.}

^{43 Ontology, p.180. UC cii. 015. 25 September 1814.}

^{44 'But if substances themselves be the subject of the division, and for the designation of the two branches of the division the words perceptible and inferential are employed, it is to corporeal substances that the characteristic and differential attributive perceptible can not but be applied, the term inferential being thereupon employed for the designation of incorporeal ones.’ (Ontology, p.180. UC cii. 015. 25 September 1814.)}

^{45 'Suppose the non-existence of corporeal substances - of any hard corporeal substance that stands opposite to you - make this supposition and as soon as you have made it, act upon it, pain, the perception of pain, will at once bear witness against you, and be your punishment, your condign punishment. Suppose the non-existence of the above-mentioned inferential incorporeal substances, ... act upon it accordingly ... no such counter-evidence, no such immediate punishment, will follow.' (Ontology, p.182. UC cii. 015. 25 September 1814.)}

^{46 Ontology, p.182. UC cii. 015. 25 September 1814.}

^{47 Ontology, pp.170, 172. UC cii. 013. 26 September 1814.}

^{48 Bentham continues by raising an objection the force of which he makes no attempt to assess: ‘But in this latter case comes an objection, viz. that the relations which on that occasion are in question, [are] - not place itself or places themselves, but such relations as belong to place.’ (Ontology, p.104. UC cii. 045. 28 September 1814.) In short, the objection is that these relations are external in regard to ‘place itself’. They are qualities of place, rather than constituents of it.}

^{49 W V Quine, ‘Things and their Place in Theories’ (hereafter, ‘Things’), in his Theories and Things (Cambridge, Mass., & London: Belknap, 1981) (hereafter, Theories), p.3. Also, Harrison, 1983, pp.64-74.}

^{50 W V Quine, ‘Five Milestones of Empiricism’ (hereafter, ‘Five Milestones’) in Theories, p.68.}

^{51 Quine, ‘Five Milestones’, pp.68-70.}

^{52 Quine, ‘Things’, p.21.}

^{53 Quine, ‘Five Milestones’, p.69.}

^{54 Jackson, 1998, pp.493-494.}

^{55 Hart, 1982, pp.10, 11.}

^{56 Hart, 1982, p.43. One of the earliest and most influential accounts of this area of Bentham’s thought is John Wisdom, Interpretation and Analysis in Relation to Bentham’s Theory of Definition (London: Routledge and Kegan Paul, 1931).}

^{57 Hart, 1982, p.9.}

^{58 Bernard Jackson appears to conflate the metaphysical and linguistic phases of Bentham’s theory in his otherwise anti-realist reading (Jackson, 1998, pp.498-499).}

^{59 Bentham often employs the distinction between genus and species. Even though he does not appear to use it in precisely this case, everything written at this point in the paper (at least) is compatible with it.}

^{60 Ontology, p.100. UC cii. 043. 28 September 1814.}

^{61 Ontology, pp.102, 104. UC cii. 044-045. 28 September 1814.}

^{62 Ontology, p.100. UC cii. 043. 29 September 1814. MS correction by textual editor.}

^{63 Ontology, p.102. UC cii. 044. 28 September 1814.}

^{64 'Set' is used here in preference to 'system' because set is a far less loaded term. There is nothing to say at present that the set is internally consistent and certainly not that it is in some sense 'complete'.}

^{65 Ontology, p.96 UC cii. 038. 26 September 1814.}

^{66 Ontology, p.94 UC cii. 038. 26 September 1814; see also Ontology, p.104. UC cii. 045. 28 September 1814.}

^{67 Harrison, 1983, p.83.}

^{68 Ontology, p.110. UC cii. 048. 28 September 1814.}

^{69 Ontology, p.86, UC cii. 024. 23 September 1814.}

This page last modified 2 April, 2009 by [Irena Nicoll]