Martii Koskenniemi has been the most influential exponent of the view that propositions of international law cannot be both objective and normative. He has famously argued that the quest for legal objectivity saddles international lawyers with the following insoluble dilemma:

The gist of Koskenniemi's argument can be represented in two pairs of general statements. The first pair reflects some of the most central aspirations of international law; the second pair reflects the reasons why these aspirations cannot be fulfilled.

1. – Propositions of international law must be capable of being objectively true or false.

2. – They must also be normative, stating what international agents ought to do rather than merely describing what they do.

Koskenniemi then claims that these aspirations cannot be simultaneously fulfilled. That is so because

1. – Propositions of international law derive their normative character from some evaluative judgements (about the right or the good) as to why certain historical facts ought to constrain the behaviour of international agents.

2. – Evaluative judgements are not capable of being objectively true or false.

In short, what prevents propositions of international law from being both objective and normative is the fact that these propositions feature an evaluative – and therefore subjective – element. Suppose that I state two propositions, ‘Mary hit John’ and ‘what Mary did was morally wrong’. If the evaluative proposition expressed in the second statement cannot have a truth-value (i.e. it cannot be true or false), their conjunction in the statement ‘Mary hit John and what she did was morally wrong’ will also not admit of a truth-value.Footnote ² Similarly, if value judgements cannot be true or false, any theory that makes the determination of international law depend even partly on value judgements will be unable to generate true or false statements about its subject.Footnote ³ This is not to say that such statements will be altogether useless, since they may produce a helpful report of the speakers’ feelings or their interests.Footnote ⁴ But that would be small consolation, since neither moral nor legal statements purport to report feelings or interests. I am not saying ‘what Mary did made me feel angry’, or ‘what Mary did made me feel sorry for John’. My statement purports to reveal what is the case about Mary's action – that is, that it is wrong – in just the same way that statements of international law purport to convey what is the case about international law – for example, that it allows or prohibits certain conduct, rather than the speaker's interests or feelings about what the law should be. If the sceptics are right, however, the ambition of everyday moral and legal propositions to state what is the case rather than what the speaker wants or feels is incoherent and implausible.

Koskenniemi's argument and the sceptical challenge it articulates are not specific to international law.Footnote ⁵ If values are subjective, then any theory of law, national or international, that assigns values a central role will have to face the sceptical challenge. There are, of course, some reasons to think that the sceptical challenge may be all the more potent in the international case. It might, for example, be argued that evaluative debates in national communities are tempered by the fact of a strong sense of shared history and common essential values, a sense that could plausibly be said to be lacking – or to be much more diffuse – in the case of the international community. I do not venture a thesis as to whether this argument is correct. It is enough for my present purposes that the sceptical challenge confronts all lawyers who are concerned with the objectivity of their domain, including international lawyers, whether or not that challenge might be especially threatening to the domain of international law. So although I shall mainly conduct my discussion with international law in mind, the thrust of my thesis will track the more global theoretical ambitions of Koskenniemi's argument on the point.

There are, I think, two plausible strategies for defending the objectivity of international law against the sceptical challenge. The first strategy would begin by accepting the main sceptical claim that any attempt to produce an account of international law that is both objective and normative is doomed to fail. It would then make its way out of the sceptical challenge by abandoning or relaxing the claim to normativity.Footnote ⁶ In practical terms, this strategy would favour a formalist approach to international law, according to which propositions of international law may be objectively valid or invalid depending on whether they can be properly inferred from certain valid premises, but they do not generally create any moral rights and obligations for international agents.Footnote ⁷ The second strategy would confront the sceptical claim directly. It would hold that objectivity and normativity are compatible and that the ambition to produce an account of international law that displays both features is not philosophically incoherent.

My aim in this paper is to show that there exists at least one version of this second strategy that can succeed against the sceptical challenge. I shall argue that scepticism about the objectivity of evaluative judgements is incoherent and that the opposition between the objectivity and the normativity of international law is illusory. My argument will proceed in four stages. First, in order to provide some context and to set some stakes for the ensuing discussion, I shall try to put the sceptical challenge in the best light, to outline its main supporting intuitions (section 1) and to connect the scepticism of international lawyers with deeper philosophical accounts of value-scepticism (section 2). I shall then distinguish between different kinds of scepticism about values in order to focus attention on the particular kind of scepticism that supports the claim that objectivity and normativity are incompatible (section 3) and draw on resources from the philosophy of mind and language to show that this kind of scepticism about values cannot be coherently formulated (section 4). Finally, I shall take up the objection that, even if propositions of international law can be both objective and normative, to talk of legal objectivity is pointless because it can have no operational impact on our debates about the correct interpretation of international law (section 5).

The article does not stake a position about whether the first strategy against the sceptical challenge could possibly succeed – that is, whether there can be a plausible theory of international law that makes no claim to normativity.Footnote ⁸ The most I can insist on in the present context is that insofar as value-scepticism is incoherent, it furnishes no argument why those who care to defend the objectivity of international law should necessarily be formalists in general or positivists in particular. If value-scepticism cannot get off the ground, then – for all their other potential attractions – formalism or positivism cannot be preferred over rival theories on the sole ground that they defend the law's objectivity against sceptical attacks. Equally, my article does not provide any argument to show that theories of international law that assign a central role to normative judgements (e.g. policy-oriented or natural-law theories) are preferable. My aim is simply to show that such theories cannot be dismissed as contenders for providing the best account of international law on the sole ground that they assign a central role to evaluative judgements. Finally, although I shall devote considerable attention to Koskenniemi's work, I focus on his arguments only as an example of some more widely shared perceptions about the objectivity of value. If anything, I believe that the rejection of value-scepticism can be useful in turning our attention from the metaphysical claims in Koskenniemi's thesis to the political case he makes against liberal conceptions of international law (I return to this point in the conclusion).

More generally, in showing why value-scepticism must be dismissed, the paper's more limited ambition is to throw new light on and to recast the debate between formalist and non-formalist theories of international law. If the objectivity of statements of international law is guaranteed against scepticism whether or not such statements include an evaluative element, the choice between formalist theories of international law (such as positivism) and their rivals must turn on other, substantive grounds, such as on the ability of each group of theories to fit our ordinary thinking about what international legal practices require and to provide an attractive account of the values that underlie those practices. Of course, we can expect the precise content of this substantive test for choosing between formalism and its rivals to be a matter of some controversy. My point is that having stopped worrying about sceptical threats to the objectivity of our claims, we can devote all our energies to resolving the substantive disagreements that give these theoretical debates their vitality and practical significance.

1. Some attractions of scepticism about the objectivity of values

I start from the idea that, as international lawyers, we need to employ evaluative judgements in several aspects of our work. Most obviously, we often need to assess whether the state of a certain part of international law is satisfactory – for example, whether it lives up to the demands of international justice or fairness or whether it displays a certain bias or prejudice. Sometimes the law itself requires interpreters to engage in some evaluation – for example, when it asks us to apply standards of reasonableness or equity. But evaluation is also central in determining what the law says in the first place. The facts of international practice never speak for themselves; one always needs to pick out the facts that are important or salient for the purposes of international law from the heap of factual information that might be available; in other words, we need what John Finnis called a ‘criterion of significance’.Footnote ⁹ This is how we can tell that a state representative's official statements can have legal consequences for the state she represents, but that the colour of her shirt cannot. This last point is not uncontroversial, especially in positivist quarters, but that controversy only concerns the extent of the lawyer's engagement in evaluative judgement (i.e. whether we also need to make such judgements in finding out what the law is). It does not dispute the need to use values either as the moral measure of positive law or in whatever ways positive law itself instructs us to.

We have several things to worry about when we are called to make evaluative judgements, much more so when these judgements pertain to an agent's rights and duties. We worry about the impartiality of our judgements and we take it very seriously when someone accuses us of bias. We worry about the reasonableness of our judgements and take it seriously when one disputes the validity of the reasons that support them. We worry about the scope and sharpness of our judgements and pause for thought when someone doubts their correctness in a particular case or their ability to give a determinate answer to a particular problem. All these questions are practical. They raise challenges that we face every day. At the same time, they are questions that we know how to deal with. A charge of partiality calls on us to re-examine some of our background assumptions for bias. A charge of unreasonableness calls on us to look at the reasons supporting the objector's viewpoint. A charge of narrow-mindedness calls on us to look carefully into the case at hand and reaffirm or qualify our original judgement. We know how to deal with all these charges because we have a good grasp of, first, how to check whether they are well-founded and, second, how to correct ourselves if they are.

The concept of objectivity is flexible enough semantically to describe any of those practical concerns about impartiality, reasonableness, and sharpness and it is probably these concerns that lawyers and theorists have in mind when they think of objectivity as an ideal for which the law ought to strive.Footnote ¹⁰ But we can also use objectivity as a placeholder for a very different set of concerns, which are explicitly philosophical in character. Here are some: that our evaluative judgements only express what we feel; that our judgements cannot be true or false independently of the fact that we happen to believe that they are true or false; that our judgements are not made true by any feature of the world.Footnote ¹¹ If true, any of these philosophical claims could support the conclusion that our judgements are not ‘objective’. Furthermore, their validity would be an inescapable fact of our condition: whereas we may fruitfully strive to make our judgements more impartial or sharper, if it is essentially true of our value judgements that they can only reflect our feelings or personal interests, then there is nothing we can do except accept our predicament. The issue I want to turn to is whether we have reason to worry about these philosophical claims about evaluative objectivity, over and above the practical concerns of the previous paragraph.

For a start, it is difficult to resist the thought that appeals to ‘objective truth’ are metaphysically suspect, insofar as they seem to ask us to believe in a wholly counterintuitive universal ontology, which includes ‘things’ called values among the real entities.

Furthermore, trying to prove the objectivity of our evaluative judgements seems self-defeating and pointless. You might value friendship, whereas I value solitude. You might think that abortion is morally permissible, whereas I think that it is always immoral. You may think that the invasion of Iraq in 2003 was justified, whereas I think that it was a horrific moral blunder. Clearly, we can debate these matters a lot and perhaps reach a point where we agree or become unable to articulate our disagreement any further. What is much less clear is the point of discussing whether my or your views on these matters are somehow ‘objectively’ correct or true. Do we have reason to think that there is some ‘real’ world of values to which we can appeal in order to find out who is right? If nothing like an observable ‘moral universe’ exists, and if a discussion of objectivity involves no more than a repetition of our original arguments, then talk of objectivity of values might even be counterproductive. It might give us an excuse to stick to our beliefs even in the face of a better argument, or, worse, it might make us believe that we are right and others are wrong without having to defend our position with serious, rational argument. Is it not better to acknowledge that reasonable people may differ in what they value and that there is little point in trying to dismiss some of their judgements as ‘objectively mistaken’ and to promote some others (usually our own) as ‘objectively true’?

Third, some form of subjectivism or scepticism about the objectivity of values draws on the powerful idea that each of us has to decide what to value and what not to value for ourselves. When it comes to deciding such questions, other people have no right to force their view of right and wrong on us, to indoctrinate us, to make decisions in our stead, or to undercut our deliberations on the ground that they know better. Indeed, the burden of making value judgements for ourselves is a feature that we recognize as constitutive of our ethical and moral agency.Footnote ¹² This is an aspect of our freedom that we want to use with critical intelligence and sensitivity, but which we are not prepared to give up in favour of any luminary (usually a religious or political leader) who claims to have ‘objectively true’ moral or ethical insights. The demand for toleration and respect for difference also follows from this conception of agency. Respect for others includes the recognition that they are equally capable of carrying their own burdens of judgement and that in doing so they might well reach conclusions different from our own. As Barry Stroud has put it,

The argument acquires added bite when the frame of reference becomes international. Different communities and different cultures will adopt their own moral and ethical codes of conduct and one may reasonably expect those codes to diverge in important respects. Talk of ‘objective’ international values might seem calculated to induce an impression of a global uniformity that flies in the face of moral and ethical diversity across different cultures.

Having said that, this opposition to the idea that value judgements might be objective does not sit comfortably with the way we normally think about evaluation and disagreements on matters of value. For one thing, value-scepticism entails that people cannot be mistaken in their value judgements. It suggests that we may find some views on matters of value horrible, detestable, or insensitive but we have no ground for saying that the people who hold them have made a mistake. By the same token, scepticism entails that our disagreements about, say, the morality of the invasion of Iraq or the permissibility of abortion have no subject matter, for there is nothing that we actually disagree about.Footnote ¹⁴ If our evaluative judgements are no more than reports of our subjective feelings or emotions, then our debates about matters of value amount to no more than exchanges of reports of our respective psychological states, with nothing to share or divide between them. Indeed, if the sceptics are right, the very idea of evaluative agreement or disagreement is inherently confused, for both of these attitudes require some issue, a case on which people's views may be said to converge or differ.Footnote ¹⁵ A further implication of value-scepticism is that evaluative arguments can never be interesting in themselves, for such arguments only tell us things about the speaker, not about the way things are. Unlike statements of fact, the sceptic says, judgements of value can never simply ‘be’ true or false. Rather, they are made true or false by the person of the speaker; the proposition ‘the invasion of Iraq was immoral’ can be true for you or for me, but false for the British government, for Halliburton or for George W. Bush. Its truth-value changes depending on the identity of the person who utters it. This way of thinking about values and evaluative disagreements has significant consequences for our conversational ethics. On the one hand, it leads us to look at evaluative disagreements as unbridgeable voids, which can be overcome only by leaps of faith and empathy, not by rational debate (given that there is nothing to be debated).Footnote ¹⁶ On the other hand, it tempts us to reduce our evaluative debates to disagreements about what are supposed to be the only matters that admit of correct and mistaken answers, namely matters of fact.

These points highlight the counterintuitive streak in value-scepticism and the suspicion that it may misrepresent what goes on in normal moral and ethical deliberation. We tend to think that people can be mistaken in saying that an action was wrong or right and that they are generally able to identify and correct those mistakes through rational argument. We also tend to think that our disagreements about values are genuine in the sense that there is always a case that we are trying to defend or dismiss. Moreover, we think that the worth of an argument does not generally depend on the identity of the person who expresses it. If a person said that international law is prejudiced against women and someone else responded ‘well, she would say that because she is a woman’, we would find the response both unfair and beside the point. We would ask the respondent to state his case against the argument itself, not against the speaker. Finally, we think that our evaluative disagreements are not wholly reducible to disagreements about some matter of fact. People can have all the facts straight about the status of an embryo and still disagree vehemently about whether abortion is morally permissible.

These points suggest that we have reason to be concerned about the implications of value-subjectivism for the character of justification to others in instances of evaluative disagreement. The reason is that the truth of value-scepticism might effectively deny us the possibility of rational evaluative argument and interpersonally valid justification. If we disagree, can we justifiably think that the other party is in error? Can we appeal to some fact of the matter about, say, moral right and wrong to settle our disagreements? If not, what guarantee do we have that our arguments and justifications do actually hold up? At the same time, the suggestion that we should reject value-scepticism on the ground that its endorsement would make it difficult for us to sustain our standard picture of moral deliberation is not convincing on its own. The sceptic could reasonably object that the standard picture is wrong and it would be better to get rid of it. Perhaps there are no genuine moral and ethical disagreements, but only some moral or ethical positions that we share and some that we do not. Similarly, perhaps it is futile to insist that all our evaluative disputes are rationally resolvable. The mere fact that the truth of these points would have momentous consequences for our moral thought is not necessarily a reason to regard their scepticism as false.

In the following sections I shall assume that the truth of value-scepticism is an open question. My aim will be, first, to identify the main arguments in its favour and sharpen the claims that these arguments support and, second, to show why the sceptical intuitions are not just compatible with the idea of evaluative objectivity but also depend on that idea for their intelligibility. In a nutshell, I shall submit that the reason why value-scepticism fails is not that its truth would have dramatic consequences for our moral thought, but that its truth would entail that the sceptical position could not be coherently formulated in the first place. In particular, drawing on the work of Donald Davidson, I shall argue that sceptics such as Koskenniemi ignore the crucial link between objective truth and the meaning of propositional content. A domain of thought bears the mark of objectivity when judgements or propositions within that domain are intelligible – that is, when interpreters can typically understand their meaning. If propositions within a domain are intelligible, then it is an objective matter whether they are true or false. Crucially, intelligibility is a basic commitment of the sceptic, since it applies to her own doubts. To be capable of asking for a justification, the sceptic must assume that both her position and that of her opponents are intelligible. She must therefore concede that it is an objective matter who is right and who is wrong. If this is correct, then contrary to the sceptic's assumptions, one cannot claim to understand what an evaluative judgement claims and still deny that the judgement can be objectively true or false.

2. Two arguments against the objectivity of value: Kelsen – Koskenniemi – Mackie

The key idea of value-scepticism is that judgements of value are a matter of subjective belief or preference and that, consequently, they do not admit of a truth-value. This idea features prominently in some of the most important works on international legal theory. Hans Kelsen was one of the foremost advocates of subjectivism about political values. One of the arguments that Kelsen uses in order to defend his positivist pure theory of law is that judgements of value do not admit of truth or falsity and therefore cannot feature in a scientific theory of law. He writes,

A similar form of subjectivism has made its presence felt most sharply in the most comprehensive and influential work on modern international legal theory, Martti Koskenniemi's From Apology to Utopia. Koskenniemi's position on the question of the objectivity of values can be summarized in the following statement:

A statement of this view can also be found in Koskenniemi's critique of the idea that international justice might play a role in the interpretation of customary international law. He says,

Koskenniemi's scepticism towards the objectivity of values occupies a central position in his general argumentative strategy. He claims that international legal discourse, which he regards as dominated by a liberal ideal of distinguishing law from politics, is condemned to an endless oscillation between appeals to historical argument about what states have done, believed, or accepted, and appeals to international values.Footnote ²¹ The reason for this oscillation is that although appeals to the history of international practice (treaties, customary practices, etc.) need to be justified by reference to some point, purpose, or value that these practices serve, the attempt to identify that point, purpose, or value runs counter to the fact that matters of value do not admit of a right or wrong answer. If this is correct, every effort to turn the legal project into more than an abject historiography by infusing it with evaluative content is condemned to lack objectivity. Koskenniemi is not alone in this assessment. Critical theorists such as David KennedyFootnote ²² and Anthony CartyFootnote ²³ share a similar view, despite their other differences with Koskenniemi's own account. For the sake of simplicity, my discussion in the rest of the paper will focus on Koskenniemi's arguments, noting these parallel claims and differences of approach where necessary.

I should note at the outset that the peculiar strategy of Koskenniemi's argument creates a problem of attribution. Koskenniemi does not advocate an outright value-subjectivism, but rather wants to attribute that position to liberal political thought. His arguments for subjectivism are therefore always couched in terms of what liberals ‘would say’ or ‘could consistently claim’, rather than what Koskenniemi himself takes to be the correct view.Footnote ²⁴ This strategy gives rise to three questions: first, whether Koskenniemi is right to think that liberals are value-sceptics;Footnote ²⁵ second, what his own position on the matter is;Footnote ²⁶ and, third, whether value-scepticism as described by Koskenniemi is a coherent position, even if liberals or Koskenniemi do not actually endorse it. My concern here will be with the third question only and it is in that light that I shall treat the arguments Koskenniemi offers for value-subjectivism as his own.

Koskenniemi's critique of evaluative objectivity consists of two strands of argument. The first strand premises the subjectivity of values on the rejection of certain ontological accounts about the place of values in the world. Discussing the dissatisfaction of modern theories of international law with the idea of a ‘natural normative order’, the truth of which is independent of what international agents think or believe, Koskenniemi writes,

The thrust of this strand in Koskenniemi's argument is that accepting the objectivity of values would commit us to a weird ontology, which would need to make room for a ‘divine will’ or some other kind of ‘natural order’, which ‘pre-exists States’ and forms a ‘closed normative code’. The fact that we find such ontological claims highly implausible gives us reason to doubt whether judgements of value can have objective truth. Koskenniemi states the point even more simply in his conclusion:

The second strand in Koskenniemi's critique of evaluative objectivity relies on the fact that evaluative judgements seem to be essentially contestable. This point emerges clearly in his discussion of the modern interpretation of early communitarian international legal thinking:

The argument here is not focused on the dubious ontology of objective values but on the fact that international lawyers can have deep disagreements that reach all the way down to their most fundamental respective assumptions about the origin and content of the right and the good. The essential contestability of those assumptions (or the incompatibility of these ‘conceptual schemes’Footnote ³⁰) casts serious doubt on the idea that any of the competing views about the good and the right could lay a claim to objective truth.

To appreciate the radical character of Koskenniemi's sceptical claims, note their difference from the more modest claim that the truth of evaluative judgements is relative. Claims of relativity do not dispute that evaluative judgements can have a truth-value; they only suggest – often very plausibly – that the truth or falsity of a given value judgement should depend on the context in which the judgement is expressed – for example, putting up one's feet may be acceptable behaviour in one community but quite offensive in another. This means that, just like the objectivist views that Koskenniemi is attacking, claims of relativity too are only intelligible on the assumption that evaluative judgements admit of truth or falsity.Footnote ³¹ Their ‘relativity’ concerns the situation, not the speaker.

The arguments from the ontological weirdness of objective values and their essential contestability have found support not just among international lawyers but also among moral and ethical philosophers. In the rest of this section and the next two I want to link the scepticism of international theorists like Koskenniemi with some general philosophical accounts of value-scepticism. I believe that looking into these philosophical arguments will not only help us to appreciate the full force behind the claims of sceptical international lawyers but also to understand those claims as part of a broader and deep-rooted philosophical tradition.

One of the best-known advocates of the attack on evaluative objectivity is John Mackie, whose Ethics: Inventing Right and Wrong presents the two arguments that underlie Koskenniemi's value-scepticism in their most general philosophical form. Mackie formulates the argument from the essential contestability of evaluative judgements (which he refers to the as ‘the argument from relativity’) in the following cautious way:

Mackie then turns to the ontological argument (which he refers to as ‘the argument from queerness’) and makes the following claim:

Since their first appearance, Mackie's arguments have continued to challenge ethicists and they have elicited a variety of philosophical responses.Footnote ³⁴ By contrast, Koskenniemi's scepticism has gone largely unchallenged in international legal circles. In the following section I shall try to identify some of the assumptions that lend force to value-scepticism and to draw attention to the reasons why certain kinds of response to the sceptical arguments are unlikely to succeed. This discussion will pave the way for a more accurate statement of the problem of evaluative objectivity.

3. Deconstructing the sceptical case (I): internal versus external scepticism

When trying to state the problem of evaluative objectivity in the right way, it is important to appreciate that scepticism about values comes in two varieties. Consider an example. Two sceptics happen to overhear you and me disagreeing about an article in the newspaper which claims that the music of the Spice Girls is better than Wagner's. One of them says, ‘These chaps are both mistaken. They are having an evaluative disagreement and we know that there is no objective answer as to which view is right and which is wrong on matters of value.’ The other responds, ‘My friend, you are falling for the same trick as they are! By claiming that their views are false, you have already conceded that their evaluative judgements admit of a truth-value. If you want to maintain your sceptical stance, as I suppose you do, all you can say is that your view simply differs from theirs, that you feel differently about different sorts of music, and perhaps add that your view is preferable because you have studied musicology, you hate opera, and so on. What you cannot do is say that your view is objectively true.’ Ronald Dworkin has aptly called the first kind of scepticism internal and the second external.Footnote ³⁵ The difference between them is that internal scepticism claims the same logical space, and requires the same kind of support as the statements whose truth it is disputing. Internal sceptics, such as the first interlocutor, need to appeal to the concept of objective evaluative truth as much as their rivals do. Thus although they may pose a challenge to the substantive claims about what is truly the case, they pose no challenge to the objectivists’ claim that the substantive question admits of true and false answers. By contrast, external sceptics consider themselves wholly liberated from the task of disputing the truth of anyone's evaluative claims.Footnote ³⁶ Their simple but crucial claim is that we ought to guard against the folly of thinking that our preferences or beliefs are anything more than just our preferences and beliefs. They urge us to argue for, defend, promote, or revise our evaluative judgements and to reject, criticize, or accept those of others; but they warn us that we are not entitled to claim objective truth or falsity for either.

It is easy to see why sceptics about the objectivity of values will always try to align themselves with the external perspective, whose distinguishing mark is that it rejects the notion of objective truth about values without losing the intuition that we can have meaningful disagreements about our evaluative preferences and beliefs. Indeed, both Koskenniemi and Mackie want to maintain their attack on evaluative objectivity while making clear that they have their own firm (although not true!) views on moral and political issues.Footnote ³⁷

So leaving the toothless challenge of the internal sceptics to one side, we need to ask what exactly one would need to demonstrate in order to rebut the external sceptical claim that evaluative propositions cannot have a truth-value. It might seem that there are two probanda here. First, one would need to provide reasons for us to accept a world ontology that, contrary to our immediate experience and intuition, makes room for distinct, perceptible, and real entities called values. Second, one would need to show that our frequent and deep disagreements about matters of value should be attributed to our varying success in perceiving those elements of reality, rather than to our different subjective desires or ways of life. However, I think that the two sceptical arguments are at core alternative formulations of the same assumption. They both assume that proving the objectivity of evaluative judgements requires one to show that values are real, that they exist somewhere in the ‘external’ world and that we possess a special, although of course fallible, faculty for sensing their presence in the way we sense the existence of celestial bodies, coins, or vacuum cleaners. This is true of Mackie's and Koskenniemi's argument from the contestability of evaluative judgements. The argument gains its force from the assumption that the only way in which we could arbitrate our disagreements about what is good, right, or beautiful would involve a belief in the existence of distinct and real evaluative entities, a bare appeal to which would clinch the result for one or the other competing claim. For Mackie and Koskenniemi, our evaluative judgements are essentially contestable exactly because there exists nothing in the real world that could show them to be objectively right or wrong. The source of their scepticism about values, then, lies in the ontological argument.

Some philosophers (typically referred to as ‘realists’) have thought it possible to confront and defeat the sceptics’ ontological argument on its home ground, insisting that values exist as real features of the world.Footnote ³⁸ I cannot here discuss the full merits of a realist response to a sceptic about the objectivity of values. I believe that if we are offered an alternative that shows subjectivism about values to be misconceived, without getting into questions of what ‘exists’ in the world or committing us to a grand ontological scheme, we should take it. Furthermore, it seems to me that the question of evaluative objectivity does not necessarily require us to enter the ontological debate. If you are an objectivist about values, the only idea you need to defend is that the truth-value of an evaluative proposition does not depend on the identity of the speaker. Insofar as you manage to defend this claim without appealing to values as part of the furniture of the world, you will be entitled to dismiss Koskenniemi's portrayal of objective values as ‘closed normative codes’ which ‘pre-exist States’ and reflect some sort of ‘divine will’ or ‘natural order of the world’ as unnecessary ontological clutter.

Having said that, asking why both the sceptic and the realist feel so strongly committed to the idea that proving the objectivity of value judgements must be an ontological enterprise can help us sharpen the question of evaluative objectivity a little more. I would suggest that the appeal of the ontological enterprise lies in two fundamental intuitions that, as both realists and sceptics appreciate, any plausible theory should be able to meet. The first intuition is that a theory of evaluative judgements ought to make ample room for the distinction between what we believe and what is the case.Footnote ³⁹ The second, apparently conflicting, intuition is that a theory should show the truth of our evaluative judgements to be connected to our beliefs and sensibilities, or it will end up making it completely inaccessible to us, something reserved for the intelligence of gods and omniscient beings.Footnote ⁴⁰ Whilst making sure that there is space between truth and belief, a theory should not define the truth of our evaluations in a way that makes it wholly independent of our ‘conceptual schemes’ and our beliefs about the world. The question is whether we can make sense of the idea that our beliefs on matters of value can be objectively true or false without talk of grand ontological schemes and without sacrificing the intuition that moral, ethical, or aesthetic truth cannot be identified independently of what we believe.

4. Deconstructing the sceptical case (II): truth, objectivity, and radical interpretation

Drawing on the work of one of the foremost analytic philosophers of the last century, Donald Davidson, in this section I try to show that we do not need to think of the objectivity or truth-aptness of evaluative judgements in realist or ontological terms. Rather, the objectivity of evaluative judgements is entailed from the structure of our beliefs and our ability to understand and interpret any kind of linguistic utterance.

The argument can be stated very briefly in the following way. Sceptics assume that they can understand the meaning of a person’ s evaluative judgements and still be able to claim that these propositions do not admit of truth or falsity. This view rests on a misconception of what is involved in having a belief (indeed any propositional attitude) or understanding someone else's. To be able to understand a proposition, whether one's own or anyone else's, one must necessarily know what it would take for the belief expressed in the proposition to be objectively true or false.

I should say at the outset that I cannot go into all the details of Davidson's account of truth, meaning, and interpretation, which has been the focus of extensive discussion and research in Western analytical philosophy for the last three decades.Footnote ⁴¹ What I propose to do is to draw together some prominent strands in Davidson's thought, to highlight their interdependence, and to explain, as briefly as would befit a paper not directly concerned with the philosophy of mind and language, how they defuse the sceptical challenge to the objectivity of values.

Even those who are convinced that Descartes's cogito ergo sum fell short of demonstrating that the external world truly existsFootnote ⁴² recognize that he set out on the right track. One thing we can be sure of is that thought exists, and we have every reason to take this as a starting point for our enquiry into whatever else might exist in the world.Footnote ⁴³ One of Davidson's basic suggestions is that quite a lot follows from the existence of thought.Footnote ⁴⁴ Consider the nature of the most common vehicle for our thoughts, namely our beliefs. To start with, one cannot have a belief without understanding that it is possible that this belief may be falsified by the facts. I believe that I have five pounds in my pocket; I look and I find that I have only three. My belief turned out to be false, yet this does not make it any less true that I had it (after all, I looked into my pocket to check whether my belief was correct). The risk of falsehood, the distinction between what I believed and what was the case, was built into my belief from the start.

Another corollary of having a belief is that one cannot give one's belief any content without understanding what it would take for it to be true.Footnote ⁴⁵ To make sense of my belief that I have five pounds in my pocket, I must have a grasp of what it is that would make that belief true, namely the presence of five pounds in my pocket (or false, namely the presence of less or more than that amount). Notice that this condition holds even if I have no way of making sure that my belief is true. The statement ‘Brighton will never organize the Olympic Games’ will be perfectly intelligible to any audience, even though there is no possibility that members of that audience will be able to confirm whether the statement is true or not. The audience understands this sentence well enough insofar as it has a grasp of what would need to be the case for the statement to be true. The same point applies to our possession and use of concepts. To have a concept is to be able to classify things under it – that is, to have criteria that distinguish things that fall within the concept from things that fall outside it. Here, too, an essential part of having a concept is the awareness of the possibility of being mistaken in its application – for example, I believed that I was seeing a red object, but the object was actually orange, so my application of the concept seemed correct but was not. In all, to have concepts and beliefs, one must be able to identify the conditions that would make them or (for concepts) their application true. If thought exists, then so does the idea of truth that is independent of the content of one's beliefs about the world.Footnote ⁴⁶

Davidson extends this argument to support the idea of the holism of our beliefs. This idea says that identifying the truth-conditions of any belief presupposes that I have a grasp of a whole web of other concepts and beliefs. My believing that there are five pounds in my pocket requires that I have a grasp of the concepts of pound and pocket, as well as of basic arithmetic and spatial orientation. Given the flexibility of our use of language and sentence construction, this entails that there is no definite amount of beliefs that I must assume to be true in order to identify the truth-conditions of a given belief.Footnote ⁴⁷ Rather, to identify the truth-conditions for any one of my beliefs I must always rely on the assumption that my beliefs as a whole are objectively true.Footnote ⁴⁸ It is only against such a rich background of truths that it makes sense to think that any particular one of my beliefs or my applications of a concept may be false.Footnote ⁴⁹

The argument thus far only shows that we cannot entertain a belief or apply a concept unless we assume that our beliefs about the world are largely correct. But sceptics will justifiably demand more. They will ask how we can be sure that our body of beliefs is actually true, that is how we can be sure that our beliefs and conceptual schemes ‘latch on’ to the world as it really is.Footnote ⁵⁰ Davidson begins his response to this challenge by looking into the structure of the sceptics’ question. To formulate their doubts, sceptics must assume that they have true knowledge of the content of their own thoughts and, to the extent that they purport to understand them, the content of other people's thoughts. What the sceptics are doubting is whether knowledge of one's own mind, or knowledge of other people's minds, is enough to guarantee that one also has knowledge of what the world is like. This third variety of knowledge, the sceptics claim, does not follow from the first two.Footnote ⁵¹

Koskenniemi makes this claim at several junctures of his argument. Note, for example, one of the reasons he gives for the view that consent cannot be the basis of international legal obligation:

Koskenniemi concludes that, since such penetration of another person's mind is impossible, any attempt to show that a person's current view of the meaning of his or her past conduct is mistaken is bound to fail or, worse, to be exposed as dictatorial and unjustified.

Davidson notes that the sceptics make a crucial assumption here. They believe that our knowledge of our own minds enjoys some sort of especially secure status because of its sheer immediacy, and that it is the same status that needs to – but presumably cannot – be extended to our knowledge of the world.Footnote ⁵³ Put simply, the sceptics assume that we can ‘look inside’ ourselves, but doubt whether this tells us anything about the possibility of ‘looking out’ into the external world.

The sceptics’ assumption that an enquiry into our knowledge of the world should begin or follow from the knowledge of our own minds is false. To demonstrate this, Davidson develops a well-known theme in twentieth-century philosophy, namely the linguistic character of thought and the public nature of language.Footnote ⁵⁴ The only creatures that we can credit with thoughts are creatures that possess beliefs, desires, concepts, and so on – in short, creatures that possess what Davidson calls ‘propositional attitudes’.Footnote ⁵⁵ In turn, the expression of propositional attitudes or thoughts requires the mastery of a language. What is especially important about the linguistic character of thought is that the mastery of any language is a social, or inter-subjective, matter.Footnote ⁵⁶ The only way to learn a language is to become apt in aligning one's verbal responses to certain aspects of the world with those of one's fellow creatures – for example, to say ‘table’ when one sees a table, or to say ‘five pounds’ when one sees five pounds. In fact, as Wittgenstein's famous rejection of the ‘private language’ argument has demonstrated, the social or inter-subjective nature of language is exactly what allows one to make the distinction between true and false applications of words and concepts, because it provides the only possible criterion of checking the identity of one's responses to a given aspect of the world.Footnote ⁵⁷ Confirmation that I am using the word ‘table’ correctly when I am looking at some thing can only come from the fact that users of English would display the same verbal behaviour when looking at the same thing.Footnote ⁵⁸ Without this background of shared responses to a shared sensory stimulus (here, seeing a table), I could have no concept of my use of language being correct or mistaken, and thus I could have no thoughts. The linguistic character of all thought entails that the sceptics’ assumption was mistaken: knowledge of one's own mind is not a privileged pivot from which any attempt to know the content of other minds and of the world must depart.Footnote ⁵⁹ Without the knowledge of other minds that is involved in the inter-subjective character of language, one could not give one's own thoughts any content whatsoever.

Davidson then extends the thesis about the linguistic character of thought to explain how, against the sceptics’ assumptions, the mastery of a language, which is what grounds the knowledge of what is inside the mind, is possible only through knowledge of the world. Following the lead of W. V. Quine,Footnote ⁶⁰ Davidson approaches the matter by asking the following question: given that knowledge of any mind (one's own or another's) depends on the existence of a shared language, what conditions need to obtain for an interpreter to be able to interpret with success the verbal utterances of a speaker? In order to put the question of interpretation in its sharpest or most radical form (hence the reference to his theory as one of radical interpretation), Davidson supposes that interpreter and speaker do not share the same language and asks what conditions would need to obtain for the former to interpret successfully the utterances of the latter. By discussing the problem of interpretation in this very difficult context, Davidson undertakes to lay bare the kind of knowledge that is required for all linguistic understanding, without relying on any prior knowledge of the speaker's language, beliefs, or conceptual scheme.

Davidson argues that radical interpretation is only possible on condition that the interpreter credits the speaker with two attributes, which collectively amount to what Davidson calls the Principle of Charity in interpretation.Footnote ⁶¹ First, the interpreter must credit the speaker with the same apparatus that makes the interpreter capable of having thoughts and language – that is, the interpreter must attribute to the speaker's beliefs the same degree of holism or coherence that characterizes his own body of beliefs. Second, the interpreter must credit the speaker with the same ability to respond to aspects of the world that the interpreter attributes to himself (and to other users of his language). Davidson calls the first attribute the Principle of Coherence and the second the Principle of Correspondence. He describes their effect in the following way:

Notice that the attribution by the interpreter to the speaker of basic logical consistency and responsiveness to the world may credit the speaker with more than he actually is entitled to (e.g. the speaker may be a parrot), but it is no more charitable than the attribution that the interpreter must make to himself in order to make sense of having thoughts and beliefs.

Consider the following example. Suppose that you are sharing a table with someone whose language you do not speak. At some point that person points to a five-pound note on the floor and exclaims ‘πέντɛ λίρɛς!’ Now, as long as you can see what he is pointing at, you would probably understand very quickly that his exclamation means ‘five pounds!’ or something to that effect (e.g. ‘a banknote!’, ‘you dropped something’). Why is that? Davidson's claim is that your interpretation of ‘πέντɛ λίρɛς’ as ‘five pounds’ was generally successful because you credited the speaker (of Greek, as it happens) with two attributes: first, a capacity to have concepts and beliefs in the same way as you do and, second, a like ability to respond correctly to a source of sensory stimulation, namely the existence of the fiver on the floor. In other words, the reason why your interpretation of the Greek phrase was successful lay in your assumption that the speaker's beliefs about the world were in agreement with your own and therefore, by your own standards (these are all you have), largely true.

In this way Davidson's argument demonstrates that, contrary to the assumption that gives scepticism its currency, knowledge of the content of any belief is only possible on the assumption that we, and the speakers we interpret, have a largely true view of the world. Our three varieties of knowledge are, after all, interdependent: knowing the content of our minds and of the minds of others requires us to have knowledge of the shared environment in which thought acquires its content (so the idea of a conceptual scheme which is unintelligible to us is incoherentFootnote ⁶³), while knowledge of the world is only possible for creatures that are able to have and to communicate thoughts and beliefs about it.

Notice that Davidson's argument does not entail the obviously silly conclusion that all our beliefs are correct. The attribution to the speaker of logical consistency and a largely correct view of the world is a necessary condition for radical interpretation, but the attribution of particular beliefs to the speaker is always defeasible by evidence (usually empirical) that the speaker actually holds another belief.Footnote ⁶⁴ There is enough space between the attribution to speakers of a largely right view of the world and the attribution to them of particular beliefs to accommodate interpretive error. For example, the more often I observe that the speaker utters the word ‘σκύλoς’ when – as far as I can tell – he sees a dog, the more support I have for attributing to him the belief that he sees a dog. Once I have enough evidence to treat this attribution as reasonably safe, I am entitled to treat as mistaken the speaker's use of the word ‘σκύλoς’ to refer to what, to me, is a table.Footnote ⁶⁵ But it may well be that the source of mistake lies with me; perhaps my observation of the speaker's behaviour when he says ‘σκύλoς’ was inaccurate or I may have had little chance to observe the full range of situations in which that word is used. These are instances where interpretation fails. But the important point, again, is that recognition of this failure requires a grasp of what it would have taken for interpretation to have succeeded, and it is on that very ground that we stand in order to explain error.Footnote ⁶⁶ No less than the attribution of a true belief to the speaker, the attribution of a false belief or mistake is intelligible only against the assumption that speaker and interpreter share a largely true view of the world.

5. The relevance of evaluative objectivity

We all share the concepts of good and bad, right and wrong, beautiful and ugly, useful and useless, and we make judgements using those concepts every day in our moral and legal practice. We communicate and understand each other's claims about these concepts and we frequently disagree about which claims are true and which are false. Sceptics such as Koskenniemi and Mackie argue that even though we are able to understand ourselves as being in evaluative disagreement, we cannot claim truth for our evaluative judgements. I have tried to expose this view as incoherent. To say that value judgements can be objectively true or false is not, as the sceptics assume, to postulate the existence of weird entities in the furniture of the universe. It is to say that the notion of objective truth is essential to our ability to have evaluative thoughts and beliefs and to interpret the thoughts and beliefs of others. Without the grasp of the concept of truth, no understanding and – a fortiori – no disagreement is possible, no matter whether the proposition involved is factual or evaluative. Indeed, only a shared background of agreement on true values could give our evaluative disagreements their bite and interest and explain their occasional and baffling intractability. In one of his last essays, which is worth quoting at greater than usual length, Davidson put the matter as follows:

The sceptic might want to avoid this conclusion by saying, ‘Well, if understanding the other's evaluative judgements commits me to accepting that these judgements admit of a truth-value, then I feel inclined to say that I do not understand them in the first place.’ But that trick would be too cheap: as long as we share the same language with those whose judgements we are interpreting, there is little possibility that their utterances will strike our ears as mere noise.Footnote ⁶⁹

Putting scepticism about values to one side, in this section I want to turn attention to a rather different but equally popular complaint against talk of objectivity in morality, ethics, politics, and law.Footnote ⁷⁰ The complaint, put forward most forcefully by Richard Rorty and Jeremy Waldron, might be stated as follows. It may be true that questions of value admit of right and wrong answers. Knowing this, however, does not in the least help us resolve our everyday moral, ethical, or aesthetic disagreements, since all sides to such disagreements will claim to have the true answer. The appeal to the objectivity of values will therefore drop out of our disagreement as useless, since it can offer us no ground for discovering which ethical, moral, or aesthetic view is right and which is wrong.Footnote ⁷¹

In one sense, the complaint is surely correct. Trying to attach the label ‘objectively true’ to one or the other view does nothing to reinforce it vis-à-vis its rivals.Footnote ⁷² Indeed, whenever people claim in the midst of a genuine and reasonable disagreement that their view simply reflects the truth, we tend to take that as a sign that they have run out of decent arguments for their position. Does this not entail that talk of truth and objectivity deserves no place in our practice of ethical, moral, legal, or aesthetic deliberation? Will we not get on just as well with our substantive debates without any appeal to those concepts?

We should regard this suggestion with suspicion for two reasons. The first, which proponents of the complaint in question will acknowledge, is that talk of objectivity and truth is surely necessary in order to defend our deliberative practice against certain fashionable sceptical challenges, such as those by Mackie and Koskenniemi. Sceptics about truth, as Stanley Cavell has put it, hide in the space between what can be said and what can be the case, between what can be articulated and what can be true.Footnote ⁷³ There is no other way of forcing them out of their hideout but to talk directly about the object of their doubts.

To see a second reason for insisting that truth and objectivity remain relevant in our deliberative practice about ethics, morality, or law, consider the following contrast. There are evaluative questions on which we care to distinguish right and wrong answers (e.g. when is pre-emptive self-defence justified?) and questions on which we do not (e.g. is strawberry ice cream better than vanilla?). We tend to say that the former are issues on which we should be aiming to discover the truth, or issues that require us to exercise careful judgement; we say that the latter are issues of subjective preference and taste. It should, of course, be clear by now that there is no interesting philosophical difference between the two groups of issues. The concept of objective truth is equally essential in understanding claims about both self-defence and ice cream, if these claims are intelligible at all. Rather, the claim of those who find talk of truth and objectivity superfluous in everyday moral, ethical, or legal debates is that the notions of truth and objectivity exhaust their usefulness on this very abstract philosophical level.Footnote ⁷⁴ Thinkers like Rorty and Waldron would acknowledge that the concept of truth is necessary for people to be able to understand and perhaps disagree with each other. They claim, however, that this concept tells us nothing about the disagreements themselves and thus can play no role in their development and resolution.

We have already noted that the last suggestion is surely correct. Truth and objectivity are not arguments in a performative sense. They do not clinch debates, nor do they put one of the competing views to disadvantage. Yet the claim that they are not relevant to or useful in those debates must be resisted, because it ignores the special motivating force that the conscious pursuit of truth exerts over our attitude towards a given proposition. In this connection, consider the important difference in your attitudes towards the legality of pre-emptive self-defence, on which we agree that there are objectively true and false answers, and towards a comparison of ice cream flavours, which we think of as a matter of taste. To make your case for or against pre-emptive self-defence, you feel compelled to consider the arguments of those who are seriously engaged in the same debate and you wish your arguments to be considered fully by them too. You also want your disagreement to lead all participants in the debate to reconsider their positions and to admit mistake or confusion whenever that is exposed. Now, provided that your fellow discussants are well-meaning and honest, in such a debate you will get the sense that you are all working towards a common end and directing your argumentative efforts towards the same target. Whatever the result of the debate, your feeling will be one of shared achievement or failure. By contrast, your attitude to disagreements about ice cream flavours will be very different. There you are likely to ‘let go’ of any debate more easily. You will not feel compelled to justify your view in any length, to engage deeply with other people's views on the matter or to revise your view when someone points out an inconsistency in it. In fact, you may even admit that someone else is right in order to get rid of what seems to you a pointless disagreement. This marked difference in your attitude, or your conversational ethic, towards the two issues calls for some explanation.

One could try to explain this difference without appealing to the concepts of truth and objectivity. One could attribute it to the fact that the legality of the use of force is an inordinately more important matter for us than finding the best ice cream flavour.Footnote ⁷⁵ In a different world, where people did not know how or care to use force against each other and instead believed that ice cream is the nearest thing to the gods’ ambrosia, finding the best ice cream flavour would be the more important project and people's attitudes in debates about it would resemble their present attitude towards the question of pre-emptive self-defence. Any talk of truth and objectivity in such debates can only be parasitic on identifying the project that interests people more.

However, the appeal to the notion of importance is simply too thin to explain the nature of the difference in our attitudes towards projects in which truth is at stake and other projects. In particular, it tells us nothing about the reasons that move us to take the distinctive attitude that we do towards a topic whenever we understand that the purpose of debating it is to discover the truth. For we may agree that we seek truth only when we think that it is important to do so, but we still need to explain how the judgement that a certain topic is important generates the special attitude that we adopt when trying to discover the truth about it. In short, we want to ask why the pursuit of truth on some topic moves us to adopt the distinctive conversational ethic that we do when debating it.

Here is one explanation. The concept of truth does not just give sense to our utterances; it also moves us to recognize the intersubjective nature of our quest for right answers.Footnote ⁷⁶ When participants in a debate recognize that the aim of their effort is the discovery of truth, they are moved to see that they stand a better chance of getting close to it by allowing their individual perspectives to engage and fertilize each other. Debates genuinely focused on discovering truth are characteristically open and constructive in exactly this fashion. They resemble less a group of perspectives ‘battling it out’ until one of them dominates than a process of constant reformulation of each perspective in the light of the other until a mutually satisfactory equilibrium is reached.Footnote ⁷⁷ This is an attitude we cannot even begin to account for in (the obviously self-referential) terms of the ‘importance’ of the topic in question for us. It is an attitude that we need to attribute to the special motivational force that the pursuit of truth exerts on human beings and the special conversational ethic that constitutes its distinguishing mark. If this is correct, talking about truth and objectivity in the context of our ethical, moral, or legal debates may not offer an argument for one or the other of the competing views, but it points to the constructive attitude or ethic that participants in a truth-oriented debate need to adopt if their joint effort is to bear fruit.

6. Conclusion: modesty and charity, not subjectivism

This article has tried to show that the supposed opposition between the normativity and the objectivity of international law is no more than an illusion. I have argued that there is no reason to deny truth-value to evaluative judgements or, for that matter, to any theory of international law that accords a central role to such judgements. If value-laden theories of international law are indeed inferior to theories that exclude appeals to evaluative judgements, that will be so for other, substantive reasons. Laying value-scepticism to rest affords us a clearer view of the ground where the genuine theoretical battles between formalists and their opponents must be fought.

But I think that one would be justified in arguing for a broader point. Taking a stand on matters of value is often perceived as abandoning the sphere of rational debate and retreating into the world of indemonstrable conviction. Sceptics like Koskenniemi supply a philosophical story about truth that supposedly explains and supports this attitude. They say that we are not assertive about our evaluative beliefs – and we are suspicious towards those who are – because matters of value do not admit of right or wrong answers. Consider the following example of Koskenniemi's argument against objective values:

This claim seems to me an odd mixture of scepticism and an uncharitable reading of liberal politics.Footnote ⁷⁹ On the one hand, it should be clear by now that the sceptical story about ‘knowing better’ one's mind or having to ‘penetrate the subjectivity’ of another in order to understand the meaning of their utterances is not credible. If meanings and correct interpretations are not ‘in the mind’ to begin with, scepticism about values cannot sustain any argument about the privacy of evaluative thought. On the other hand, and more importantly, we can explain the core intuitions that lend Koskenniemi's scepticism its surface appeal without falling for the incoherent philosophical claim that values are subjective, a matter of taste and preference, the only way to respect other people's views and the like. In fact, we have at hand two far more plausible stories to account for our non-assertiveness and our liberalism about questions of value. The first story is theoretical. We can accept that all questions of value do admit of true and false answers and still say that, on at least some matters, agents ought to be free to discover those answers for themselves (all liberal political philosophers take such a position in respect of questions about the good). The second story is more practical. We need to remember that even though they admit of true or false answers, disagreements about values are sometimes very difficult to define with precision, let alone arbitrate. This is a fact of life that we must register not just in theory but also in the way we conduct our evaluative disagreements about the point and purpose of international legal practices. In fact, the very nature and orientation of our international legal and political debates require that we adopt an open, critical and charitable conversational ethic. The availability of right and wrong answers, and our occasional conviction that we are right and others are wrong, may entitle us to dismiss scepticism about values as incoherent, but it does not give us licence to lay aside the virtues of modesty, charity and circumspection.