In The Shadows: Conservative Epistemology and Ideological Value

This article intervenes in the debate about the nature of conservatism. Some contributors to this debate have claimed that this ideology can be defined as an adjectival disposition. They claim, that is, that a conservative possesses an attitude towards shared values rather than a distinct set of substantive values. The following discussion interrogates this account of conservatism and concludes that it can only be coherent if we ignore the epistemological limits of conservative thinking.

arise when we attempt to define conservatism in epistemological terms. In turn, it will claim that an adjectival definition of conservatism can only be valid if we ignore a crucial aspect of conservatism; its inability to construct a conception of value within its own epistemic boundaries. The concluding section then explores 'nominal' and 'substantive' definitions of conservatism and rejects them in favour of a dialectical one.

Adjectival Conservatism
Some accounts (Huntington, 1957) have described conservatism as an ideology that has no consistent values. Of these, Kieron's O'Hara's study is perhaps the most expansive. In it, O'Hara defends the notion that conservatism is 'adjectival' in nature. He defends, that is, the notion that conservatism is a commentary on shared values rather than a set of values itself. At the centre of his argument is an epistemological definition of conservatism. In contrast to Robin and others, O'Hara (2011: 24) suggests that the conservative is, above all else, the advocate of a sceptical epistemology. This epistemology is expressed in the form of two propositions: the 'knowledge principle' and the 'change principle' ['kp + cp']. These principles are described as follows: [kp] Because society and its mediating institutions are highly complex and dynamic with natures that are constantly evolving as they are co-constituted with the individuals who are their members, both data and theories about society are highly uncertain (O'Hara, 2011: 49-50).
[cp] Because the current state of society is typically undervalued, and because the effects of social innovations cannot be fully known in advance, then social change (a) must always risk destroying beneficial institutions and norms and (b) cannot be guaranteed to achieve the aims for which it was implemented. It therefore follows that societies should be risk-averse with respect to social change, and the burden of proof placed on the innovator, not his or her opponents. It also follows that change, when it does come, should ideally be incremental, (b) reversible where possible and (c) rigorously evaluated before the next incremental step (O'Hara, 2011: 88). These two principles can be employed to establish an ideological orientation that we might regard as being distinctively conservative. They can also be reconciled with the views of many notable conservatives who have sought to define conservatism. Consider, for instance, the accounts of conservatism offered by Michael Oakeshott and Roger Scruton. Both of these writers (Oakeshott, 1991: 423;Scruton, 2006: vii) challenged the idea that conservatism comprised fixed beliefs and values, and both rooted their conservatism in epistemological claims about the dangers of rationalism. On his part, Oakeshott made a deliberate attempt to empty conservatism of substantive ideological content. Although he acknowledged that particular ideas and assumptions that had come to be associated with English conservatism, he was anxious to prize them apart from conservative reasoning. In his view (Oakeshott, 1991: 423), the conservative disposition was 'not . . . necessarily connected with any particular beliefs about the universe, about the world in general or about human conduct in general'. It was instead a particular orientation towards governing and governance that began from understanding 'ourselves as we have come to be' (Oakeshott, 1991: 424). It is perhaps unsurprising that when he came to sum up conservatism, Oakeshott (1991: 408-409), he associated it with a set of dispositions rather than substantive beliefs: To be conservative . . . is to prefer the familiar to the unknown, to prefer the tried to the untried, fact to mystery, the actual to the possible, the limited to the unbounded, the near to the distant, the sufficient to the superabundant, the convenient to the perfect, present laughter to utopian bliss (Oakeshott, 1991: 408-409). This passage, which closely resembles O'Hara's change principle, seems to establish some clear desiderata that the conservative can employ. Problems arise, however, when we attempt to identify the reason why the conservative may prefer the 'familiar to the unknown'. O'Hara claims that this preference can follow from adjectival reasoning. He suggests that the knowledge principle and change principles constitute the basis for a commentary on shared values rather than a 'substantive' set of claims about value. But it could be claimed (Beckstein, 2015) that this is an insufficient basis for conservative political thinking. After all, it is possible to claim that the conservative cannot make any assessments about the status quo unless they have established some assessment of its moral, emotional or functional value. O'Hara responds to this line of reasoning by introducing another clause to his description that concerns the concept of 'public reason'. Conservatives, he claims, will seek to preserve whatever is valued by the society in which they inhabit. They will do so not in order to preserve an abstract social ideal. Rather, they will do so in the name of achieving 'equilibrium' (O'Hara, 2016: 436). 2 The below discussion considers O'Hara's three principles and explores the problems we encounter when we use them to construct an epistemological definition of conservatism. The objective is not to discount O'Hara's logic but to acknowledge that this logic reveals the limitations of conservative reasoning if we define it in epistemological terms. The article concludes by offering a historicist explanation of these limitations.

The Knowledge Principle
We can begin by considering the knowledge principle. Ostensibly, this principle provides a guide to the conservative when they are faced with rationalist claims about political and social life. If all beliefs that we have about the world are uncertain, it follows that we should be sceptical about truth claims. This might induce a cautious handling of certain kinds of information and a reluctance to accept abstract theories. The knowledge principle is, however, 'empty' unless it is accompanied by judgements that are non-epistemological in nature. Here, we can make two points in relation to O'Hara's knowledge principle. First, we must acknowledge that an ontological clause is featured in O'Hara's exposition of this principle. The value of the epistemological clause ('theories about the society are highly uncertain') is dependent upon the claim that it makes about the 'highly complex and dynamic' nature of society. If we follow its logic, we might assume that if a society emerged that was simpler and more intelligible, the knowledge principle would be dismissed by even the most ardent conservative. 3 In short, the conservative would need to abandon the knowledge principle in light of observations they had made about their social reality. This thought experiment exposes the problem with adjectival definitions of conservatism. For it reveals that adjectival conservatism cannot supply its own justification. The knowledge principle can only stipulate that the social order is not easily understood. It cannot provide an argument for why the social world is not easily understood. To construct such an argument, the conservative needs to reach beyond the knowledge principle to make substantive claims about the world in which they live. This requires reasoning that is not epistemological but ontological.
A conservative might respond to the above claim by drawing attention to the epistemic role that tradition plays in guiding conservative thought. A conservative, they might claim, will be more likely to trust knowledge that has been 'proven' through experience and observation (Dorey, 2010: 31-40). In doing so, they could argue that there is a body of knowledge that can be supplied to make value judgements that are conservative in nature. But this notion threatens to contravene the ontological clause that is featured in the knowledge principle. As O'Hara (2011: 50) himself claims, 'a theory's success in accounting for past data does not mean that it will be successful in its predictions of future observations'. If this argument is accepted, it follows that knowledge that applies at one moment will not necessarily apply in another. There is, therefore, no reason why an advocate of kp + cp should trust 'past' or 'traditional' knowledge any more than they trust other forms of knowledge. We can, of course, conceive of arguments that the conservative could make about the value of 'practical' or 'traditional' knowledge. Michael Oakeshott based much of his political philosophy upon such claims. 4 But these arguments would be based on claims about value that the non-conservative may not share. It would be difficult, then, to regard the conservative who makes them as the bearer of an adjectival ideology.
In a sense, then, the knowledge principle is 'empty' unless it is fleshed out with ontological observations. But these ontological observations do not necessarily have to comply with the knowledge principle. They are theories about the world. And if the conservative takes them to be first-order components of their thinking that are regarded as immutable, they are violating the notion that our theories about the world are uncertain.
When we consult the major statements of conservative thought, we frequently encounter writers who employ beliefs about the nature of the world as a starting-point for their conception of value. Take, for instance, Russell Kirk (Kirk, 2015: 31), who arrived at a conservative position on authority by exploring the consequences of the 'irremediably flawed' nature of the human character. The conservative political theorist Stanley Parry (2015) went further. In his efforts to reconcile reason with tradition, he questioned whether it was sufficient to appeal to the concept of 'natural law' to oppose progressive rationalism. His (Parry, 2015: 155) reasoning is worth quoting: [T]he conservative appeal to reason within the natural-law tradition fails to cope with the problem in its existential form. Its basic error is to appeal to nature as the source of order, precisely when ill minds perceive nature as the source of disorder, as the dilemma from which they must save themselves. A simple counter-assertion cannot work the therapy needed. The real problem is to move to a perception of nature as ordered by a transcendent purpose, whose intention can be learned only by a revelation from on high. The real solution is to move from the threat of disordered nature to the perception of the right order that has been determined by divine intervention.
In essence, Parry escaped from the shadows of reason by appealing to a higher form of knowledge that it could not comprehend. For the purposes of this article, Kirk and Parry's respective arguments are revealing in two respects. First, they demonstrate that when the conservative attempts to provide a justification for their values, they tend to reach beyond the knowledge and change principles to make substantive claims about either the nature of the world or the divine forces that have shaped it. Second, they demonstrate that once a conservative has made such claims, they begin to arrive at arguments that are substantive rather than adjectival. If we disagree with them, we cannot, in the last instance, be accommodated within the conservative tradition. We might share their proponents' sceptical view of change and question the virtues of rationalism. But we will not satisfy the requisite desiderata until we have embraced particular views about the character of human nature and its relationship with the transcendent. In the case of Parry, it is difficult to see how he could condone a conservative defence of tradition that did not appeal to the idea of a transcendent God.

The Change Principle
Recall that the change principle places a burden of proof upon the innovator and that it establishes the universal value of incremental, as opposed to rapid, change. It is possible to see how this principle would provide a guide to political decision-making. Yet, it can only do so when the conservative makes value judgements about the nature of change. In the first instance, the conservative must assess what constitutes significant innovation. After all, human actions of all kinds constitute change of some kind. Even the act of acquiring knowledge through perception involves a change in an individual's relationship with the social world. Some criteria are, therefore, required to distinguish those changes which are significant and those which are not. 5 As we shall see later, the search for these criteria must entail enquiry beyond the confines of the knowledge and change principles. For the time being, however, our attention can be directed towards the dilemma the conservative faces when they apply the change principle after a change has taken place.
Let us consider Edmund Burke, who O'Hara frequently cites to illustrate conservative reasoning. 6 Burke (1967: 35) rejected the French revolution because he regarded it as an episode of change that had destroyed beneficial institutions. In turn, he wanted to restore the pre-revolutionary order. He was thus an innovator. And as an innovator, he would, according to the change principle, be held to a greater burden of proof than his revolutionary enemies who had created the status quo. As O'Hara (2016: 431) states, the knowledge and change principles dictate that 'the positive aspects of the status quo are threatened by innovation, and its negative aspects will not necessarily be corrected by the planned innovation'. It is difficult to see how Burke could have justified innovation without contravening O'Hara's principle. He might have claimed that innovation could be justified given the inadequacy of the post-revolutionary order. But in order to do so he needed to claim that the society he was seeking to recreate would be better than the one established by the revolution. He would need to establish, that is, that the risk of change was worth taking. That could only be done by appealing to the idea that an imagined order could be better than the existing order. That imagined order may well have existed in the past. Yet, its value could only be established by making judgements about its qualities. It could not be justified on the grounds that either it (a) would not risk the destruction of beneficial institutions and norms or (b) that it would be guaranteed to achieve the aims for which it was implemented.
O'Hara is, of course, correct when he claims that the conservative's ability to endorse change is informed by their conception of risk. But in order for an actor to assess risk, they must make value judgements about the status quo. We can return here to Oakeshott (1991: 411), who regarded change as an event that entailed 'certain loss and possible gain'. If a conservative believes that the existing order is bad, they will be more likely to recommend change in pursuit of a potential gain. Conversely, if they believe the existing order is good, they will be suspicious of any innovation and will emphasise the loss that is likely to follow. As the conservative makes value judgements about the status quo, they are necessarily required to engage in forms of thought that might have a little relation to either the knowledge principle or the change principle. Note that when O'Hara enunciates his change principle, he states that change may pose a risk to 'beneficial institutions'. Here, we might pose a question: how does the conservative determine whether an institution is beneficial? O'Hara anticipates this question by claiming that when the conservative makes value judgements about the status quo, they make reference to the knowledge and change principles. When he (O'Hara, 2011: 107-108) considers the conservative view of authority, for instance, he suggests that this view is informed by epistemological concerns. Conservatives, he claims, tend to respect authoritative individuals and institutions if they have endured on the grounds that they reproduce 'practical knowledge'. As he puts it: Authoritative figures and institutions that have survived are likely to have some value for a society -their mere survival is a prima facie indication of their functionality [. . .] Authority which has passed the test of time is also a better guide [. . .] to wise or moral behaviour.
But it can be questioned whether the knowledge or change principles sanction authority in this way. The knowledge principle dictates that theories about the world are uncertain. It does not suggest that the longevity of a particular body of knowledge is a guarantee of its value. We can establish a hypothetical scenario to establish this point. Consider, then, a socialist state that has survived for several centuries. When they occupy a society governed by this state, the actor who was wedded to the knowledge principle could not condone the status quo on the grounds that the state's authority was derived from socialist -and thus rationalist -theories about the world. And in a society governed by such a state, the change principle would dictate that the burden of proof would be on the innovator who wished to change it.
O'Hara (2011: 108) qualifies his claim about the conservative disposition towards authority by suggesting that the conservative will only endorse 'legitimate' forms of authority. Yet, this authority cannot be determined without making recourse to some kind of criterion that cannot be supplied by the knowledge and change principles that O'Hara sketches. The conservative will need to judge, for instance, whether democratic arrangements produce good forms of authority. Significantly, these judgements will be needed to make the knowledge and change principles operational. They are not second-order judgements that can be separated, at a logical level, from kp + cp. Even if we could discover a logical relationship between the knowledge and change principles and a particular conception of authority, this relationship would undermine the notion that conservatism is adjectival. That is because this relationship imbues conservatism with substantive values about authority. Put simply, if we recognise that the change and knowledge principles lend themselves to a certain idea about legitimate authority, we must accept that an actor who holds different views about authority is, in some sense at least, less conservative than they could be. Again, the adjectival quality of conservatism appears to be difficult to sustain.
At some junctures, O'Hara also threatens to associate conservatism with values and concepts that appear to have no necessary relationship with the knowledge and change principles. One such concept is that of stability. When, for instance, O'Hara (2016: 434) attempts to dismiss the notion that adjectival conservatism is preoccupied with 'normative risk', he introduces the notion that the conservative will seek a society 'stable enough to allow him to pursue his idea of the good'. This notion seems to be indebted to Oakeshott's conception of 'disequilibrium', and it is employed to legitimate a certain kind of value pluralism. Indeed, stability is conceived as a prerequisite for the reconciliation of different values. Provided that it is maintained, individuals can be free to express their own values without impeding the expression of others. But to suggest that stability is a good thing is to necessarily make a value judgement. If the conservative was to occupy a society in which they believed, on the basis of their rational enquiries, that it was not present, they would feel compelled to rectify that absence and propose change. In such a context, defenders of instability would possess a value that could not be reconciled with conservatism, yet, because instability would then constitute the status quo, they would, in accordance with the change principle, be subjected to a less rigorous burden of proof than their conservative opponents.
Similar reasoning presents itself when we consider other indicators that the conservative may employ to assess the health of a society. O'Hara (2011: 102-103) claims that there are 'social indicators', such as levels of poverty and violence, that can be employed to determine whether a society is functioning 'in a reasonably objective and non-relative sense'. It is difficult, however, to regard such indicators as being, in any sense, objective. If the conservative inhabited a society that had always suffered from high levels of poverty, they would have no point of reference outside of that experience. They could thus argue that poverty is immutable and that any attempt to eradicate it would entail certain loss and possible gain. The judgement they make would be a relative, not objective, one. Importantly, these values cannot be derived from the knowledge principle or the change principle. They require the conservative to assess the history of the societies they occupy and to make substantive value judgements about what a good society would look like.
O'Hara does threaten to accept this argument when he draws attention to the values that conservative may hold regardless of their (epistemological) conservatism (O'Hara, 2016: 437): Instead of conservatism being the qualifier of ends-based ideologies -a conservative liberal, a conservative green -it becomes the mark of any practical ideology, after which one becomes free to pursue one's own ideas of the good. One can, like Burke, be a liberal conservative, or like Scruton, a green conservative.
The key phrase here is 'after which'. This phrase invites us to regard the application of the knowledge and change principles as separate 'moments' in the conservative thought process that precede the consideration of substantive values. The problem with this notion, as the above discussion has argued, is that these principles cannot function without reference to 'ideas of the good'. That is, in part, because they cannot supply their own justification. They can only do so when they are brought into a dialogue with substantive beliefs and values.
This point relates to a claim made by Beckstein (2015: 18) in his discussion of conservative ideology. The adjectival conservative's commitment to the change principle might compel them to be wary of significant reform. But they will have values that may correspond with those of other ideologies. Their conservatism is thus accidental. If they could establish that the risk associated with the change had been reduced, they could justify change on the same grounds as their progressive opponents.
O'Hara might respond to this claim by suggesting that such 'ideas of the good' can be supplied by another conservative resource, namely that of 'public reason'. But as we will see, this resource will not necessarily supply the knowledge and change principles with any justification. O'Hara (2016: 424) appeals to the concept of 'public reason' to support his claim that conservatism is adjectival in nature. A conservative's values, he claims, must be 'culture bound' because they must be defensible within the society that she occupies. The conservative, he argues, is not a relativist because they must ensure that their ideas and values are amenable to public reason. Not all members of a society will accept the conservative's arguments. These arguments should, however, be defensible within the parameters of a society's modes of thought. The upshot of this notion is that (O'Hara, 2016: 429) the conservative will be tied to the values that their society endorses:

Public Reason
If kp+cp is to support the case for conservatism as adjectival, then it cannot entail any particular value . . . The values a conservative will wish to preserve . . . will depend on what a particular society or culture valorises.
Problems arise when we identify public reason as a guide for the conservative's value judgements. We must first consider a hypothetical situation in which a particular society rejects the conservative's epistemological reasoning. In such a situation, the conservative could no longer appeal to public reason in support of the knowledge principle. Nor could they endorse the change principle. For they would need to endorse change in order to inhabit a society in which they could once again appeal to the principle of public reason. Imagine, for instance, a society in which Rawlsian epistemological ideas are widely endorsed. The conservative would be compelled to preserve those ideas and the institutions that reproduced them, even though they contravene the knowledge principle. The conservative might, of course, claim that such a scenario is unlikely based on their claims about, say, human nature. But the conservative would no longer be making an adjectival argument; they would instead be making a substantive one about human nature and its consequences. An individual with different values about human nature could not necessarily be a conservative.
The second problem concerns the concept of public reason and its relationship with the knowledge principle. This problem can be expressed in the form of a question: how can the conservative be certain that they have an adequate knowledge of public reason? If they are, as O'Hara's definition stipulates, epistemological sceptics who are suspicious of theories about the world, they should be careful when they develop a description of collective values and beliefs. Such a description will necessarily be abstract in nature. It will also require adjustment as a society's values changed, even if such change was rapid and frequent.
A final problem presents itself when we consider the reasons why particular actors might hold particular values. O'Hara (2016: 430) claims that an adjectival conservative can endorse all manner of values. 'Any conservative', he writes, 'is such that there exists a value that he holds, but it is not the case that there is a value such that all conservatives hold it'. That may well be true. But those values have no necessary relationship with the knowledge and change principles. We may even claim that any actor who is attached to particular moral or ideological values will be reasoning in an unconservative way when they express an attachment to those values. Such values can only come into being when the actor makes judgements about the quality of either the status quo or an imagined past or future. They cannot emerge spontaneously from the knowledge and change principles. And they may, in certain circumstances, threaten those principles. An actor who values liberty necessarily possesses a 'vision of another, different and better world' (Oakeshott, 1991: 428), if they live in a society in which they do not believe that those values are present. To advocate those values, they might reason in all manner of ways that contravene kp + cp. We can carry this notion forward by seizing upon a statement O'Hara makes about Burke when he defends the adjectival nature of conservatism: [Burke's] faith did not suggest to him the universality of Christianity, but rather the importance for a society of its religion.
Here, O'Hara is attempting to use Burke to demonstrate that a conservative can possess, and indeed sanction, all kinds of beliefs. But it is not necessarily the case that this example proves such an argument. Burke's belief in Christianity may not have led him to advocate the dissolution of alternative traditions of religious belief. It did, however, result in a preference for the preservation of religious belief per se. Significantly, his defence of religion could not established by the knowledge and change principles. Some argument would need to be made for the virtues of religious belief on the basis of its existence value, its moral value or its function.
It is not difficult to imagine a scenario in which the conservative would find it difficult to maintain the knowledge and change principles. Consider once again a conservative who occupies a society that has embraced a Rawlsian conception of justice. Because Rawls' epistemological ideas contradict the knowledge principle, this way of thinking about justice will contravene the essence of conservatism. But the conservative will be compelled to accept it unless they make a substantive argument about the nature of social justice. This substantial argument cannot be derived from the knowledge and change principles. At best, the former would result in a suspicion of Rawls' epistemological claims. But the change principle would place the burden of proof on the innovator to demonstrate than an alternative conception justice would produce better outcomes. Kp + cp say nothing about the merit or otherwise of the status quo. They do suggest the status quo is 'typically undervalued'. But because they do not specify what criteria we might employ to determine value, they are 'empty' until they are brought into contact with substantive arguments. These substantive arguments, whether they are moral, philosophical or otherwise, will have an independent value that could gain first-order status within the ideological morphology of conservatism.
When they searched for beliefs and principles that they associate with their disposition, some conservatives have threatened to expose this feature of conservatism. We can consider, for instance, the description of conservatism that the late Roger Scruton (2015) offered in How To Be A Conservative. This book defines conservatism through negation. Each chapter considers an ideology and identifies a 'truth' within it that conservatives can accommodate. Scruton (2015) claims, for instance, that for all of its evils, socialism does contain an essential truth: the principle of mutual dependence. Scruton's objective, of course, was to elevate conservatism above its progressive rivals. 7 The reader is invited to conclude that whereas progressives, in their pursuit of utopian perfection, ignore the truths of history, conservatism can accommodate and reconcile those truths. But a sceptical reader might question Scruton's method. In his efforts to extract kernels of truth from the ideologies that the conservative must negate, Scruton effectively admitted that conservatism can only supply its own justification by reaching beyond its own resources of reason. He exposed, that is, the conservative's futile quest to escape from the shadow of reason. The conservative might oppose socialism in the name of liberty, but they must then defend authority in opposition to liberalism.
A conservative must plunder its opponents' ideologies for what James Alexander has helpfully termed 'shadow ideals'. These ideals do not have the same status as those that are present in progressive ideologies. If they did, the conservative would contravene O'Hara's knowledge principle by upholding them. 8 They are instead imitations of those which oppose the logic of whoever might be advocating change. They do not follow from a rational argument about the nature of the good society. Rather, they emerge from an opposition to the progressive's rational arguments that is both incomplete and contingent. That is, in part, because it is an opposition that depends on its dialectic relationship with the other. If the object that is casting a shadow disappears, so does the shadow itself.

Alternative Definitions: Nominal and Substantive Conservatism
Once we have acknowledged the difficulties of establishing a conservative philosophy of value that has an epistemological foundation, we might look elsewhere to sustain an adequate definition of conservatism. We might, for instance, identify a conservative attitude towards the status quo as the starting-point for the conservative assessment of value. This would be to construct a nominal, as opposed to an adjectival, definition of the ideology. Conservatives, it can be argued, value what is rather than what is not (Cohen, 2011). This value does not follow from any particular emotional, aesthetic or ethical disposition. Rather, it follows from an attachment to the familiar by virtue of its existence. There is an insufficient space to comment on these kinds of claims in considerable detail. It is nonetheless necessary to situate them within the broader debate about the nature of conservatism. The central argument, in essence, is that the conservative adopts a particular attitude towards value. Unlike the non-conservative, who may be willing to destroy something if the potential consequence is the creation of something with greater utility, the conservative will be attached to arrangements or practices because they have an intrinsic value. This way of thinking about conservatism does specify a calculation about value that we could regard is conservative in nature. It can be questioned, however, whether it provides a sufficiently strong desideratum for the conservative to satisfy. For, as Cohen (2011: 208) acknowledged, anyone who does not seek to control or modify all aspects of the existing order is conservative 'to some degree'. Even a utilitarian will accept the basic logic that the existing order has value if that society is organised in a utility-maximising way. Cohen's argument, of course, was that the conservative has a particular attitude towards value. In his view, the conservative values existing things more than someone who is not a conservative. I am not concerned with intervening in the philosophical justification of conservation that Cohen offered. But it is necessary to consider whether the disposition to conserve things of value is a satisfactory desideratum for the conservative. Here, we can draw attention to two problems. First, we can consider a point which Cohen (2011: 13-14) himself admitted; that if a conservative seeks to reverse even the most recent of changes, she is destroying something about the existing order that may be of value in order to establish something else. That 'something else' might have existed in the past. But it will, at the point of reaction, be only an imaginary arrangement. To attempt to realise it would surely involve a violation of the change principle and would compel the conservative to value what is not rather than what it.
There is another problem which we might consider, and that concerns the concept of disvalue. Cohen (2011: 224) claimed that it was possible to be conservative while also possessing no bias towards towards the status quo. He also argued that a commitment to destroying inequality could be reconciled with the conservative impulse. This was because he regarded intrinsic disvalue as an appropriate justification for action: Wanting to conserve what has value does not imply wanting to conserve exploitation and injustice, since they lack value. Wanting to conserve what has value is consistent with wanting to destroy disvalue.
The problem with this formulation is that it awards the conservative disposition a second-order status. Cohen (2011: 225) even acknowledged that injustice may have a certain kind of value and that the egalitarian defence of equality would need to 'trump' the value of an unequal status quo. It is difficult to avoid the conclusion that his conception of conservative value is one that is so minimalist that it fails to account for what is distinctive about conservatism. One way to repair the ambiguity of Cohen's formulation would be to strengthen its logic. We would need to replace it with one that privileged the familiar and the existing not for their particular value but for their own sake. We would, that is, need to associate conservatism with a status quo bias. But doing so would introduce another problem, namely the conservative inability to oppose change once it has happened. In the 'dynamic' world that O'Hara's knowledge principle describes, the nominal conservative would need to value whatever exists, regardless of its qualities.
The conservative can appeal to particular concepts to justify their attachment to practices, people and institutions. Some, for instance, have gestured towards the idea that the conservative disposition amount to a kind of love (Cohen, 2011). 9 Love, they claim, is something that does not follow from a rational calculation of something's present or future value. Rather, it stems from a non-contingent appreciation of something based on the meaning that has been attached to it. In one sense, this kind of reasoning does appear to have a distinctively conservative quality. It contravenes the idea that the value of something should be determined by anticipating its future or potential utility. But there is no reason why a non-conservative could not justify their attachment to an arrangement or practice in the same terms. A socialist, for instance, could value their old, dog-eared edition of Das Capital with missing pages over a shiny new one that is, in functional terms, more useful. They would, in a sense, be justifying their preference in conservative terms. We would be reluctant to call them conservatives because their attachment to the object might follow from a rational preference for socialist ideas. But in one crucial respect, they would be nominal conservatives.
Having identified the problems associated with the adjectival and nominal conceptions of conservative ideology, we might assume that a substantive one would fare better. After all, if adjectival conservatism cannot supply a criterion for making value judgements about the world, surely our only choice is to locate the universal truth claim that furnishes the ideology with a coherent belief system? This is the view taken by Corey Robin (2018), who claims that conservatism is a little more than a philosophy of inequality. It is not necessary, however, to retreat into the notion that conservatism is a philosophy of inequality, violence or whatever other concept we might locate at the centre of its ideological architecture. Here, two objections can be made. The first is that the contradiction at the core of conservatism is central to its nature. By claiming that conservatism is, say, a philosophy of inequality, we resolve this contradiction, but we threaten to conceal the fact that this contradiction shapes its character. When the conservative attempts to reconcile their values with an adjectival conception of conservatism, they enter into a distinctive pattern of reasoning that defines the conservative disposition.
The second objection is closely related and concerns the relationship between conservatism and other traditions of thought. If we claim that conservatives value particular concepts and arrangements, such as hierarchy and inequality, we gesture towards the idea that it has a universal character that transcends history. But as  first observed, conservatism is a situational ideology that obtains its identity in relation to its opponents. It also operates (Alexander, 2013) in a world that these opponents have shaped. 10 It must, therefore, engage with that world in terms that it cannot entirely understand, and that requires it to become conscious of itself. Samuel Coleman (1968: 252) once referred to 'ideological traditionalism' to denote a commitment to the preservation of tradition. We might claim that this commitment can only be made in terms that are unconservative. At the very moment that the actor becomes conscious of tradition, it ceases to be something that can be justified in terms of its existence value. It must also be justified in rational terms. In a sense, the conservative is always in the shadow of a world that they oppose, and they can only move out of its shadow by reasoning in an unconservative way. Bill Schwarz (1996: 101) drew attention to this aspect of conservatism with particular force. Reflecting on the way in which British conservatives have often imitated their enemies, he noted that Since the time of Burke, Toryism has pursued a self-declared check on the rationalism of the Enlightenment project by immersing itself in the vocabulary and practices of the Enlightenment. It has revered the past by forcing the pace of the modern; it has sanctified constitutionalism by breaking the rotation of party and by concocting all manner of one-party regimes.
If we adopt an adjectival definition of conservatism, we might conclude that this politics of imitation is merely a product of conservatism's situational character. We could argue, that is, that conservatives can sanction all kinds of arrangements and practices provided they are accordant with continuity (Scruton, 1980). But this argument is difficult to sustain. When they oppose their ideological opponents, conservatives must summon arguments and ideas that they cannot call their own. They must, in a crucial sense, be the mirror-image of their enemy. But importantly, they are defined by their dialectical relation with this enemy. Their ideals and values are constructed in the shadows of those held by others. The article will conclude by fleshing out this dialectical definition of conservatism.

Dialectical Conservatism
The above discussion has revealed the problems, we encounter when we attempt to use the change and knowledge principles to construct a theory of conservative value. These principles may provide a conservative with a guide to action. But this guide will only emerge once the conservative has reached beyond the boundaries of their epistemology to make substantive claims about value. When they reach beyond these boundaries, they are, in essence, contravening the knowledge and change principles by making substantive claims about value that may not be contingent. How do we account for this apparent contradiction? An answer to this question can be found by observing conservatism's history and recognising the role that this plays in its advocates' attitudes towards values.
As Alexander (2013Alexander ( , 2016 has noted, conservatives occupy a world shaped by changes that their conservative predecessors opposed. Yet, they are compelled to accept this world because they are unable to condone counter-revolutionary change. O'Hara would, of course, claim that this is incorrect. Conservatives, he claims, can advocate change if they can establish that the status quo is deficient. They might be more likely to defend the status quo. But they do not, he claims, need to possess a bias towards it. It is difficult, however, to regard this as a sufficient basis from which to define conservatism. If we accept the logic of this proposition, we must accept that the conservative's disposition towards change will have little to do with the knowledge and change principles. It will have much more to do with their values, whatever they might be. If they believe in the value of liberty and exist in an order that they regard as excessively authoritarian, they will be willing to advocate change. But they could only do so by making arguments about the benefits of such change. They could not appeal to the knowledge or change principles. If they did so, they would be compelled to place a greater burden of proof on the argument for change and could only defend libertarian change by contravening the epistemological basis of the knowledge principle. They could appeal to the concept of public reason. But as we have seen, doing so may entail reasoning that is incommensurable with kp + cp. Adjectival conservatism seeks a truth that is external to its logic (Alexander, 2013: 594). That is, at least in part, because it is a tradition of thought that is historically situated. Its advocates can only establish the value of a particular arrangement or practice by situating it within a story about the past, the present and the future. When they do, they necessarily engage in reasoning that reaches beyond the knowledge and change principles. It is this that explains the problems associated with O'Hara's description of conservatism. It is not necessarily the case that O'Hara's definition cannot account for the practical thought practices of conservatives. In their concrete decision-making, conservatives may well adhere to the knowledge and change principles. Nor must we accept Robin's (2018) claim that the conservative is always concerned with defending an abstract concept like inequality. But when conservatives appeal to the principles that O'Hara's definition privileges, they are offering an incomplete definition of their beliefs. For they are failing to acknowledge that these principles only acquire meaning when the conservative reasons in unconservative ways. That is, they must search for a truth that conservative epistemology cannot provide.
Conservatives have acknowledged this problem when they have searched for truth by making ontological claims about what Freeden (1996) terms the extra-human order. Some have located the truth in Christian theology (Hogg, 1947). Others (Kirk, 2015: 31) have appealed to the notion of immutable human difference that follows from the laws of biology. But all of these 'truths' share something in common. They all appeal to epistemic authority that is not supplied by the conservative's own epistemological resources. It is no coincidence that O' Hara (2011: 190-198) is unwilling to regard these appeals as being necessary components of conservatism's ideological landscape. In his view, the notion that conservatism is an ideology that affirms the extra-human origins of the social order violates the knowledge principle.

Conclusion
This article has explored adjectival definitions of conservatism. Such definitions, it has been argued, are deficient for two reasons. First, they can only be coherent if we ignore the limitations of conservative thought. Their advocates tend to assume that conservatives can award substantive values a 'second-order' status and that conservatism is, above all else, an epistemological ideology. But a conservative's epistemological principles can only acquire meaning once they are brought into a dialogue with these values. As a corollary, we must regard those values as being of central importance to conservative reasoning. Second, adjective definitions of conservatism are problematic because they cannot be upheld in all historical conditions. If a conservative inhabits a social order that they dislike, they will be compelled to innovate in order to change. And their argument for innovation will, according to the change principle, be subjected to a greater burden of proof than the arguments made by defenders of the status quo. Put simply, the adjectival conservatism cannot satisfy the change principle unless they sanction any change, regardless of its qualities.
If we accept the argument that conservatism cannot be defined as an adjectival ideology, we might search for an alternative definition that associates it with particular values. We could, for instance, endorse Robin's (2018) claim that conservativsm is an ideology that defends social hierarchy. Or, alternatively, we could settle on Alexander's (2013) argument that the conservative's true objective is to restore the sacral monarchy of the pre-revolutionary moment. This article has not been concerned with these understandings of conservatism. But its conclusions do have implications for how we might engage with them. In one sense, they lend weight to them. For they invite us to regard conservatism as an ideology like any other with own fixed objectives and values. But they also suggest that we should be wary of definitions of conservatism that resolve its contradictions. Conservatives' attempts to escape from the shadow of ideology may well be futile. Yet, they nonetheless tell us much about the nature of conservative thought practices. If we are too eager to regard them as smokescreens that conceal the 'true' character of conservatism, we may be tempted to resolve, rather than accommodate, the contradictions that define it.