Justifications, Exculpations, Causes: Epistemology and Our Image of Nature
Justifications, Exculpations, Causes:
Epistemology and Our Image of Nature


Our conception of nature makes a difference in the way we think about justification and its bearings on our epistemic practices. In this paper I look at how a sense of confinement that is associated to some forms of coherentism is provoked by a conception of nature. I examine how best to free ourselves from this confinement.

Table of contents

    1. There is a way of thinking about epistemic justification that holds that it dwells solely in beliefs. According to this view, any relation between what is believed and any item different from beliefs – maybe even if it is an item that could be a belief – is at most a causal relation. This view, commonly associated to Davidson's take on coherentism, can be stimulated by some of Wittgenstein's remarks concerning the irrelevance of interpretations when we follow rules (1953).

    This view on justification can make one feel challenged to explain how the contents of our beliefs relate to the world; how, for instance, they acquire their empirical significance. At first sight, beliefs could either have their content acquired through a merely causal interaction with relevant items of the world or could be left so that their content is not hooked to any item of the world other than beliefs. The former is the route that Quine often seemed to have favored as he assumed experience could be described in terms of stimuli in our nerve endings. The latter would amount to an image – with a distinguishable Davidsonian flavour to it – where states with content are somehow confined out of the reach of nature. If we want to gloss this last alternative in a Kantian framework, we can say that judgments are taken to be composed by assembling concepts that are themselves disconnected to intuitions, and therefore are empty – they could be at most formally valid. The world gets lost. It follows from all this that if justification is an affair of beliefs we can either hope to get an (intentional) connection to the world through causal relations or give up such connection altogether.

    In what follows I try to examine several questions around this issue in order to point at how our conception of nature matters for how we think of justification. I start out considering the ways in which we can make either of those two alternatives tolerable; this will lead me to question of what could be behind this apparent dilemma. This will then lead me to deal directly with the influence our thinking about nature makes on how we conceive of justification.

    2. The idea that causes can provide content to our beliefs depends on the assumption that our senses – or our nerve-ending stimuli – relate causally to elements of our thought. The connection would provide content if it enables us to point at a reliable correlation. Davidson, among others, has challenged the idea by pointing out that causal connections could fail to be describable in terms of natural laws (1970). In fact, Davidson's rejection of a scheme-content dualism (1974) can be seen as entailing a rejection of the idea that empirical contents could be in a law-like relation with anything else. If content always presents itself within a conceptual scheme, a law connecting contents and items of the world would have to guide our concept application and therefore this would respond to laws, rather than to our rational practices. This consequence of the rejection of the scheme-content dualism – or of what has been called the Given or, in Kantian terms, exercises of receptivity devoid of spontaneity – not only shows a connection between two theses held by Davidson (i.e. the rejection of the dualism and the resistance to psychophysical laws) but also makes clear how causal links between nature and our thinking cannot provide us with content. The presence of (rule-guided) concepts in any of our empirical contents prevents psychophysical laws even though they don't render impossible for contents to be causally connected to items of nature.

    If causes cannot provide content from nature to our beliefs, we seem to be left with the second, Davidsonian alternative: the realm of beliefs is epistemologically and intentionally confined. Justification, in this case, happens within the realm of beliefs and it is not about anything else. There is no intentional link between our thinking and nature; nature only causes some of our states. The idea of confinement follows from this conception of nature as constituted by a network of causes that cannot reach our thinking. Beliefs make sense only in the context of other beliefs and there is no special set of beliefs that would bring content to others by relating more closely with the rest of the world – the rest of the world is locked out. Since there is no belief that could be somehow closer to the rest of the world, there is little room for us to consider that some among our beliefs could be foundational. Hence, if the skeptic challenges the belief that I'm not a brain in a vat, there is nothing else that she is challenging – I can concede to be a brain in a vat and carry on holding my beliefs involving trees, tables and people. No doubting of a subset of my beliefs would force me to doubt all my beliefs. The skeptic is therefore exorcised as there is no strategy open to her other than carefully putting in question each of our beliefs. Such strategy for global doubt will in turn fail since in order to question some of beliefs she would have to hold, at least provisionally, some others. The cost of exorcising the skeptic, however, is high: we loose contact with nature and make a mystery out of our capacity to think about it. The sense of confinement follows not from our thinking not reaching anything other than thinkable contents but rather from a conception of nature according to which it stands outside our reach. Confinement makes justification belief-bound and beliefs indifferent to nature.

    3. McDowell (1994) attempted to see a way out of this confinement by rethinking our notions of experience and nature. He insisted that nature, conceived as a realm of causal connections, and experience, understood as belonging to nature, could not provide us with justifications for our thoughts; they could at most provide us at most with exculpations. We can only rid ourselves from blame for what we think by pointing at items of the world.

    It seems, however, that although exculpations fall short of constituting full-blown justifications they do play an intentional role. McCulloch (2002) claims that we don't need epistemological relations between thoughts and nature to assure us that the contents of our thoughts come from nature: exculpations could be enough. I suspect that if Wittgenstein is right about the way we follow rules, there could not be something like exculpations from nature for these would be already clues upon which we could build up interpretations. If I blame nature for my thinking of the tree in front of me – even though I don't assume my belief is justified merely by pointing at the tree – I could next present an interpretation of the rules of application of the concept of tree that would suffice to guide the application of the concept. In fact, I can only exculpate myself for what I think if I can point at something in the form of L:

    • (L) I think of a tree because there is a tree in front of me and whenever there is a tree in front of me (and a set of additional conditions hold) I think of a tree (maybe because anyone would).

    I believe the argument above against psychophysical laws prevent not only causes acting as justifications but also as exculpations.

    McDowell's position, however, can be still tempting. He claims that whatever causal connections could give us is not enough for our thinking to be rationally controlled by the world. If we are worried with constraints from what Kant called receptivity, we are likely to crave for a connection between our thinking and the world that is stronger than whatever causes can offer. McDowell labors to make intelligible the idea that in experience we open ourselves to the world. He holds that we gain conceptual content through exercises of receptivity. In order for conceptual content to be obtained from experience, nature itself has to be able to provide this content – McDowell then suggests on the need for a partial re-enchantment of nature where concepts and meanings as well as laws can be taken as natural. Re-enchantment, of course, is a remedy against the dualism of reason and nature and therefore alleviates the feeling of confinement by viewing us as capable to reach nature and be influenced by it in our own conceptual exercises. Nature is not something that is oblivious to the way we access it; quite the contrary, it can reveal itself only to those who have gone through a conceptual Bildung.

    One could then ask how we should face causal connections within nature. The worry would plausibly show up, for example, when we look at law-like descriptions of our cognitive activities; it seems reasonable to suppose that laws can describe at least the mechanisms that enable our thinking practices. In Bensusan (2000) I presented a model where the laws governing these mechanisms are not only (at most) a starting point for a process of scrutiny and revision but also themselves subject to scrutiny and revision. McDowell, for example in his 1996, favors a model of second nature that he reads in Aristotle's virtue ethics. He takes a natural law to be incapable to rationally compel us to any action or thinking: it cannot produce a rational obligation. It has therefore to be somehow brought to the space of reasons in order to make a difference to our contents. When they are brought to the space of reasons, we can rationally relate to them and they become part of our second nature. The emerging image of nature is one where there are both first and second nature but we live within the scope of second nature.1 Experience can therefore be described as (second) natural. Experience and nature are then rethought; we can gain contact with natural items that do not need to be devoid of anything that resembles concepts and thinkable contents to be objective.

    Re-enchantment, however, could still sound like not much more than a vague idea. How exactly, we can ask, are we to put together first and second nature, conceptual judgments – of which we have experience – and causes? Bensusan & Pinedo Garcia (2003) have recommended the notion of soft facts to ground our conception of the objects of our knowledge and thought. In a world that is both objective and thinkable, there is room for natural laws as they can be content of our thought. But we are encouraged to take a somehow Humean approach to causes and see them as no more than what is expressed in laws instead of some sort of hard fact that would constitute the bedrock of what is natural. Causes would then be understood only within our understanding of laws and those would depend on our understanding without which we could not use them either to predict or explain anything.2 If we give up the primacy of causes over our concepts and practices, we can accommodate them in nature and yet feel no urge to describe everything else in causal terms.

    4. Michael Williams (1996) takes McDowell's re-enchantment of nature to be a sophisticated form of foundationalism. He takes McDowell's recipe against confinement to be a change in our image of nature so that experience could be a conceptual contact with thinkable objects and therefore provide a foundation for our world view. The basic structure of foundationalism concerning epistemic justification is preserved. In fact, some thinkable contents act as products of passive exercise of conceptual capacities and are the ones that give empirical support to our judgments. Furthermore, some beliefs are central to our intentional connection to the world; some beliefs – like the presence of objects to our thinking – glue the world to our judgments. Because of these somehow privileged beliefs, the Davidsonian strategy against the skeptic vat brain argument would not hold. Justification – and empirical content – is thought as coming from a privileged source and ultimately from how things are. A revised conception of nature enables us to make sense of justification as being something other than merely an affair of beliefs: a source of justification is brought from nature through experience. Williams’ point is that McDowell’s way out of confinement seems to involve being grounded by nature, with a revised notion of nature.

    His revision in the notion of nature shows how much the way we think of justification and its relation to the world bears on how we take nature to be. McDowell’s re-enchantment of nature intends to make experience capable to ground our empirical thinking. However, the emerging conception of nature – where it is not only a network of causes – could perhaps be assimilated to an architecture of justification less suspicious of foundationalism. We can find our way out of a belief confinement if we exorcise the idea that our reasons are surrounded by (unintelligible) causes; if thinkable contents are not strangers in a world of causes and justification can be thoroughly world-involving without appealing to a grounding experience. But if we want to hold on to a re-enchanted conception of nature while rejecting the remnants of foundationalism we will also need to add a perhaps less palatable qualification. We would have to accept that our justifications would not themselves depend on our conception of nature for if they did, a challenge to the conception would make our justifications collapse. Notice that this does not happen in Davidson’s account: if nature turns out to be more than mere causes, most of our beliefs would still hold. Making that less palatable qualification makes our conception of justification less safe against the feeling of confinement. We can then feel compelled to conclude that a conception of nature that views it always from the point of view of our beliefs is maybe always on the brink of making us feel confined. Perhaps, then, we just have to make sure at all times that the world doesn’t get lost.


    1. Bensusan, H. 2000 “Induction and reason: Cognitive tendencies under the tribunal of experience”. Proceedings of the 23rd International Wittgenstein Symposium, 77-81.
    2. Bensusan, H. and Pinedo Garcia, M. 2003 “Soft facts: objectivity and the possibility of thought”, submitted.
    3. Davidson, D. 1970 “Mental events”, in: Essays on Actions and Events, Oxford: Clarendon Press 1980.
    4. Davidson, D. 1974 “On the very idea of a conceptual scheme”, in: Essays on Truth and Interpretation, Oxford: Clarendon Press.
    5. McCulloch, G. 2002 “Phenomenological externalism”, in: Reading McDowell, London: Routledge.
    6. McDowell, J. 1994 Mind and World, Cambridge: Harvard University Press.
    7. McDowell, J. 1996 “Two sorts of naturalism”, in: Virtues and Reasons: Philippa Foot and Moral Theory, ed. Rosalind Hursthouse, Gavin Lawrence, and Warren Quinn, Oxford: Clarendon Press..
    8. Williams, M. 1996 “Exorcism and enchantment”, Philosophical Quarterly, 46, 182, 99-109.
    9. Wittgenstein, L. 1953 Philosophical Investigations, Oxford: Blackwell.
    McDowell insists in calling himself a naturalist––adding quickly that he holds a naturalism of second nature. He resists any idea that could lead to some form of dualism where nature is not all. His position contrasts with a more Hegelian one that would understand nature as what we depart from through the achievements of spirit.
    Lack of space prevents a complete treatment of causal relations in these terms but I will briefly mention something that is sketched in Bensusan & Pinedo Garcia (2003). We could face laws as rules that could be grasped and used for prediction and explanation within our understanding practices.
    Hilan Bensusan. Date: XML TEI markup by WAB (Rune J. Falch, Heinz W. Krüger, Alois Pichler, Deirdre C.P. Smith) 2011-13. Last change 18.12.2013.
    This page is made available under the Creative Commons General Public License "Attribution, Non-Commercial, Share-Alike", version 3.0 (CCPL BY-NC-SA)


    • There are currently no refbacks.