1. Introduction
Traditionally, accounts of the nature of blame have presented necessary and/or sufficient conditions for what should count as part of our blaming practice. Such conceptual analysis type accounts are, however, vulnerable to counterexamples that show how the conditions put forth are either not necessary, or not sufficient for blame. These theories are thus forced to leave out part of what we would intuitively think of as blame, or include things we do not think of as blame. Recently, some alternative methods have been suggested: functional accounts of blame and paradigm-based explanations of blame.Footnote 1 In this paper I aim to explore whether the newly proposed methods are better suited to account for the nature of blame than traditional conceptual analysis is.
From the debate about the nature of blame it has become clear that a successful account of blame should – at least – be able to explain the following:
• How blame goes beyond merely judging blameworthy (Wallace Reference Wallace1994: 75). Put differently, this is the question of what the special ‘force’ of blame is (Hieronymi Reference Hieronymi2004). This question arises because it seems like we can judge someone to be blameworthy without actually blaming them, so we must be able to explain what sets blame apart from mere judgment.
• How blame can be both emotional and non-emotional. What sets blame apart from merely judging blameworthy could be the emotional element of it, however, it seems intuitively right that we sometimes blame without an emotional response (Shoemaker & Vargas Reference Shoemaker and Vargas2019: 583).
• How blame can be both expressed and private. It seems intuitively right that we can blame people without expressing it, as well as that we can blame ourselves, the dead, and those that are out of reach (McKenna Reference McKenna, Coats and Tognazzini2013: 121). At the same time, many accounts propose that blame is necessarily communicative.
• We need to be able to explain what sets moral blame directed at moral agents apart from blame directed at natural events or other creatures we do not consider to be morally responsible (Smith Reference Smith, Coats and Tognazzini2013: 36).Footnote 2
These criteria for an account of blame have emerged over time as a consequence of the counterexamples used to disprove various versions of conceptual analysis accounts of blame.Footnote 3 As David Shoemaker illustrates, all extant content basedFootnote 4 conceptual analysis type accounts admit prima facie of false positives or negatives (Reference Shoemaker2024: 13–20).Footnote 5 What this means is that the theories will either include phenomena we do not think of as blame, or exclude things we do think of as blame.Footnote 6
I start this paper with a brief overview of some extant functional and paradigm-based accounts of blame. These accounts have emerged with the same goal in mind: A more accurate and inclusive theory of the nature of blame. They all reach remarkably similar conclusions on what the paradigm-example of blame is and most of the accounts also agree on what the function of blame is. Where they differ, is in what steps they take to get to these conclusions.
In Section 3, I explore whether the accounts are able to answer the first three challenges above. I conclude that this depends on whether they can give us a plausible candidate for a paradigm-example, and a function, for blame. I show how the extant accounts have a problem in determining what the paradigm of blame is, and thus also what the function of blame is, without becoming circular or reverting to traditional conceptual analysis. To answer this challenge, in Section 4, I propose a new version of the functional account of blame. This account relies on the function of blame as explanatorily basic. To determine the function I look closer at the reasons we have for blaming, and what it looks like if we fail to blame. I conclude that we blame because we need to protect our own understanding of our moral worth. From this functional role we can derive a paradigm-example of blame as communicative. In Section 5, I show how this account can meet the first three desiderata mentioned above, without becoming circular or reverting to conceptual analysis. In Section 6 I answer some possible challenges to the account – such as whether it can also account for the nature of praise, and for non-moral blame – and end by showing how it can meet the last and fourth desiderata. In Section 7 I conclude.
2. New methods
In this section I present the particular methods of the functional account, and the paradigm-based approach to blame – before I go on to sketch how the extant accounts identify the function, and the paradigm. I focus mainly on one version of each account: Miranda Fricker’s (Reference Fricker2016) paradigm-based account and Victoria McGeer’s functional account (Reference McGeer, Coats and Tognazzini2013). I’ve chosen these two accounts because they were developed specifically as methodological alternatives to traditional conceptual analysis. Since my goal primarily is to establish a methodological alternative, these approaches are particularly helpful for my purposes.
2.1 The Method: Paradigm-based explanation
Even though her account is not the first of its kind, Fricker (Reference Fricker2016) is the first to spell out the method for a paradigm-based account of blame, as such.Footnote 7 Fricker argues that communicative blame is the paradigm-example of blame, and that as long as an instance of blame has a residue of the same point as the paradigm, it can be counted as part of the practice (Reference Fricker2016: 179). As an example, unexpressed blame can be seen as proper blame, as long as it has the same purpose as communicated blame.
Fricker relies on speech-act theory to identify the point of blame. Communicative blame, she argues, gains verbal expression through an illocutionary speech act and an illocutionary speech act has a clear illocutionary point (Fricker Reference Fricker2016: 172).Footnote 8 An illocutionary speech-act is characterized by the fact that it must have uptake in the hearer to be successfully performed. Fricker identifies the illocutionary point of blame as the point of blame in general, namely to make the wrongdoer feel remorse, and thus ‘fully acknowledge the moral significance of what they have done or failed to do’ (Fricker Reference Fricker2016: 173). Further, Fricker argues, the interplay of blame and remorse brings forth a common understanding of morality, or an ‘alignment of moral understanding’. This alignment is the first aspect of the point of blame (Reference Fricker2016: 173). The second aspect of the point of blame, Fricker identify by relying on what Austin called perlocutions (Reference Austin1962: 101). Perlocutions are the characteristic aims of speech-acts. Fricker identify the perlocutionary point of blame as ‘to prompt a change for the better in the behavior (inner and outer) of the wrongdoer’ (Reference Fricker2016: 173).Footnote 9
To sum up, the paradigm-based account identifies communicative blame as the paradigm-example of blame, and it derives the function of blame from this paradigm-example.Footnote 10 The function is identified as making the wrongdoer feel remorse, acknowledging the wrong in question, an alignment of moral understanding between victim and wrongdoer, and finally: a change for the better in the behavior of the wrongdoer.
2.2 The Method: Functionalism
McGeer develops an alternative to accounts of blame that focus on identifying essential features of blame (Reference McGeer, Coats and Tognazzini2013: 168). Essential features are features that are present in every instance of blame, and by virtue of this method any instance of blame that does not have the feature will not count as blame. Instead of essential features, McGeer suggests we look for ‘canonical features’. Canonical features ‘account for our interest in identifying a kind as such, even though things belonging to the kind do not invariably manifest the feature in question’ (McGeer Reference McGeer, Coats and Tognazzini2013: 168). Canonical features do not have to be present in every instance of the practice, as long as the instance of the practice has the function determined by the canonical feature it can be said to be part of the same practice.
As with the paradigm-based account, McGeer’s functionalist account opens for a greater variety of what we see as part of our blaming practice because it is not limited by essential features. If an instance of blame does not manifest the canonical feature, this is just an example of how blame might malfunction, and not an example that disproves that what we are talking about is in fact blame. McGeer identify the canonical feature of blame as its emotional character: resentment, indignation, and anger (Reference McGeer, Coats and Tognazzini2013: 172). This can be understood as the paradigm-example of blame on her account, while the function is both to register the wrong done, and to prompt a change in behavior in the wrongdoer (McGeer Reference McGeer, Coats and Tognazzini2013: 172).
2.3 The function and the paradigm
We have now seen how the functional and paradigm-based accounts aim to be more inclusive than traditional conceptual analysis type accounts are. In this sub-section I will look at the connections between the two accounts. McGeer’s functional account can be seen as part of a group of functional accounts connecting blame to anger and protest.Footnote 11 On her account, blame has both a backward-looking appraisal dimension – coding other’s behavior as offensively transgressive – and a forward-looking regulative dimension – aggressive, punitive, response aimed at changing or inhibiting the behavior coded (Reference McGeer, Coats and Tognazzini2013: 172). As such, the paradigm-example of blame on this account is both emotional and communicative.
On a similar, and even more recent functional account, David Shoemaker and Manuel Vargas (Reference Shoemaker and Vargas2019), identify the function of blame as signaling commitment to, and policing, moral norms. Shoemaker and Vargas argue, however, that signaling differs from communication, because one can signal even if no one is around to pick up on the signal (Shoemaker & Vargas Reference Shoemaker and Vargas2019: 594). Shoemaker echoes this conviction in his recent book (Reference Shoemaker2024: 36). Explaining how signaling can take place without communication is important for Shoemaker and Vargas in order to avoid the objection that says private blame cannot be communicative: If blame equals signaling, and signaling depends on communication, their account cannot explain the existence of private blame. If we see Shoemaker and Vargas’s account of blame as a functional account, however, they don’t need to explain how signaling can be non-communicative. When blame is successful it communicates, but that doesn’t mean that private signaling, or unsuccessful signaling isn’t also an example of blame. Communication is still an important part of successful signaling. Signaling might not be angry, as is the case for the other functional accounts mentioned above, but I think it safe to regard Shoemaker and Vargas’ account as also relying on the communicative nature of blame in order to explain its function.
Paradigm-based accounts of blame also assume that blame is paradigmatically communicative. Fricker (Reference Fricker2016) and Michael McKenna (Reference McKenna, Coats and Tognazzini2013) both present accounts of blame where the paradigm-example of blame is identified as communicative blame. On McKenna’s account of blame, the paradigm-example is necessarily communicative because this is the only example of blame that can lead to the next step – of moral account – in the conversational model he posits. This step allows for the wrongdoer to offer an excuse – or an apology – and for reconciliation and reparation to take place (McKenna Reference McKenna, Coats and Tognazzini2013: 94–95).
So far, then, we can conclude that paradigm-based accounts and functional accounts have the same motivation and are almost indistinguishable methodologically. The only difference between them is whether they put their main focus on identifying the paradigm-example of blame, or on identifying the function. Regardless, the paradigm-example and the function both seem to be necessary components of the respective method. On paradigm-based accounts the choice of paradigm-example is often explained by reference to the function it has, and the function is seen as a basic explanatory element. In other words, we cannot pick out a paradigm without also having identified a function for blame. Similarly, all functional accounts rely on a paradigm of blame as – at least – incipiently communicative in nature, and they explain the function of blame in relation to this paradigm. To avoid circularity, then, a given account must decide on which of these two features to derive from the other, and how to explain the nature of the explanatorily basic one of the two. Further, the explanatorily basic element of the theory cannot be a necessary condition on blame, or else the theory will revert back to being a traditional conceptual analysis type account and again be met with false positives and negatives. I will look closer at these challenges in the next section.
3. Challenges to the functional accountFootnote 12
In this section I will start by looking at how the functional accounts fare with the three first criteria outlined in the introduction. After this I get back to the problem of how to determine the function and paradigm-example of blame.
3.1 How inclusive are functional accounts?
Most of the accounts sketched above rely on an idea of blame as connected to reactive attitudes such as anger and resentment. The first challenge to these accounts is thus: If the paradigm-example of blame relies on emotional responses, how can the account make sense of non-emotional blame? And second, given that the paradigm-example of blame has been identified as communicative by all accounts discussed above, it seems a timely question to ask how these accounts can explain our practice of blaming in private, as well as how we can blame ourselves, the dead, and those out of reach.
An initial reply to this objection is that it is based on a misunderstanding of the method at hand. The main reason for proposing this method in the first place, is exactly that it does not have to explain all of the phenomena we count as part of our blaming practices. As long as the phenomenon has the same function as the paradigm-example it can be counted as part of the same practice, even if it is not able to fulfill this function. Non-emotional blame, private blame, and self-blame, are just derivative, non-paradigmatic forms of blame.
To explain this a bit further, think again of the goal of the functional theory. If the paradigm of blame is communicated and angry, anger and communication are what McGeer calls canonical features: features that defines the function of the practice. Communication and emotion can be essential for blame to fulfill its function, but the lack of communication or emotion does not mean that blame loses its function. Having a function then, does not depend on the function actually being fulfilled. In other words, the existence of private and non-emotional blame is not a problem for the functional account. Blame can be private, or non-emotional even if the paradigm-example of blame is not.
This solution to the problem of inclusivity comes with its own problem, however. If we accept, as do – arguably – all extant accounts surveyed above, that the paradigm example of blame is communicative, we end up with an account of blame that is too inclusive. There are many possible communicative responses to wrongdoing that we would not want to classify as blame. As David O. Brink and Dana Kay Nelkin point out (Reference Brink, Nelkin, Vargas and Doris2022: 181), one might respond cautiously by asking sincerely how the act in question was intended. This kind of careful questioning of the wrongdoer does not look like blame as intended by extant theories. If we proceed by adding constrains on the theory, however, like Fricker seems to do (Reference Fricker2016: 174), it reverts to conceptual analysis.
Even more worrisome for these accounts, however, is the question of how we can know that the paradigm we have identified actually is a paradigm-example of blame. Or, how we can know that what we take to be blame’s functional role really is so. As an example, it has been argued against functional theories that identify the function of blame as protesting moral wrong, that equating blame with protest simply shift the problem over to a definition of protest. Further, it turns out to be hard to define protest without appealing to blaming attitudes (Chislenko Reference Chislenko2019). It seems that whether these accounts can actually provide a more inclusive picture of our blaming practices depends on their ability to define a plausible paradigm-example and a function for blame without vicious circularity.
The circularity in question becomes evident by asking in virtue of what the two elements of the theory are defined. As the example above shows, if we define the function of blame as protest by reference to a paradigm – for example, a blaming attitude such as anger – we need to explain how we got to the conclusion that the paradigm is constituted by those attitudes. And we can then not explain why the paradigm is constituted by those attitudes by referring to that its function is to protest. In other words, we cannot define the function by reference to the paradigm, and then define the paradigm by reference to the function. Most accounts attempt to hold one element as explanatorily basic – either the paradigm or the function – but when accounting for why this element is explanatorily basic, they rely on the other element. As an example, Fricker holds that the paradigm-example of blame is explanatorily basic, but – as discussed above – in explaining why the paradigm is communicative blame, she relies on the function of blame. The problem is, then, how we can decide on what is an explanatorily basic feature of a functional account. And remember, the explanatorily basic feature cannot be an essential feature of blame, or we will be back where we started.
In the next section I raise more detailed worries about the extant accounts way of defining a paradigm and a function for blame.
3.2 How can we define the function and the paradigm-example of blame?
First, how do we know that the paradigm-example of blame is communicative? Julia Driver (Reference Driver2016) and Brink and Nelkin (Reference Brink, Nelkin, Vargas and Doris2022) point out that there are reasons to believe that blame might not be incipiently communicative after all. Studies of development in children show that children are able to engage in forms of blame before they can master the concept of blame on a conversational level (Driver Reference Driver2016: 218). In other words, it might be the case that we learn to blame before we can communicate. If this is true, communicative blame is at least not developmentally prior to private blame. Could it instead be that blame is paradigmatically communicative because this is the most common form of blame? It is quite possible that private blame is a more common form of blame, so it also does not seem like the pervasiveness of communicative blame is what justifies the choice of paradigm. This worry also transfers to how we define the function of blame. Brink and Nelkin point to studies that suggest children learn less when exposed to external sanctions (Brink & Nelkin Reference Brink, Nelkin, Vargas and Doris2022: 184). This could mean that what the extant accounts identify as functions, are not actually what blame leads to in real life. If a common result of blame is heightened conflict, denial, persistent wrongdoing, and so on, does this mean that the extant accounts have not been describing blame, but something else? Put differently it seems like we lack a clear argument for why we should believe that the paradigm that is being identified really is the paradigmatic form of blame. If the paradigmatic form turned out to be private blame, we could not derive the kind of function the theories surveyed above derive from communicative blame. And similarly, if we don’t know what the function of blame really is, how are we supposed to derive a paradigmatic form of blame from it?
4. A new formulation of the functional account of blameFootnote 13
In this section I will develop the functional method further. The challenges I have to answer to are: 1) how to define the function and the paradigm of blame without circularity, 2) without reverting to conceptual analysis, and lastly 3) the account must be able to explain the fourth desiderata.
I will start by determining the function of blame and establish this as the explanatorily basic element of my account. I argue that a functional theory should start by determining the function because this is the safest way to ensure that the theory does not collapse into traditional conceptual analysis. This is so, because a paradigm-example will have necessary and sufficient condition baked into its description. If it has no restrictions it simply is not a paradigm-example. A functional role, on the other hand, can be manifested by many different kinds of blame. What determines the paradigm-example, I argue, is that it is able to fulfill the functional role of blame. Fulfilling the function is not necessary for something to fall under the heading of ‘blame’, but it is a way for us to understand what the paradigm of blame would look like.
The question at hand is thus how to determine the function of blame. To answer this I first want to take a closer look at the method we are dealing with. A methodological distinction that is left unexplained by all theories that use functionalism to explain the nature of blame, is about the nature of the method itself. McGeer (Reference McGeer, Coats and Tognazzini2013) and Coats & Tognazzini (Reference Coats, Tognazzini and Zalta2024) likens functional theories of blame to functional theories of mental states and properties, while Shoemaker (Reference Shoemaker2024) and Shoemaker & Vargas (Reference Shoemaker and Vargas2019) draw on functionalism the way it is used as a theory in sociology and biology. As McGeer points out, even if our blaming practices ultimately can be traced back to a more basic form through evolutionary biology and psychology, blame the way it is today, is more sophisticated and has features it did not at earlier stages of our evolutionary history (Reference McGeer, Coats and Tognazzini2013: 184). For this reason, I will rely on functionalism the way it is used in relation to mental states.
In describing the theory of functionalism about mental states, Thomas W. Polger, uses the example of a mouse trap and a diamond (Polger Reference Polger2019). Mouse traps are designed to serve a particular function, that of trapping mice. They can take almost any shape, use different means for accomplishing the goal – a trapped mouse – and they will be evaluated on the basis of how good they are at achieving this goal. Diamonds, on the other hand, are evaluated on the basis of their components, what constitutes them. Even if other objects might have the same properties as what we take to be valuable about diamonds – their texture, look and rarity – they must contain certain essential features to be diamonds – carbon crystal with specific molecular lattice structure. This example illustrates that functionalism sees mental states more like mouse traps, than like diamonds. ‘That is, what makes something a mental state is more a matter of what it does, not what it is made of’ (Polger Reference Polger2019).
It is easy to say what the function of a mouse-trap is; it is harder to identify the function of a mental state. A mental state is not a thing we can examine in the way we can a mouse trap, and its function is not consciously chosen and determined by us. The possible functions of an attitude could be many. In empirical cases, wrongdoing followed by blame could lead to apology, forgiveness, and reconciliation. It could also, however, lead to greater conflict, or simply to disassociation and severed relationships. As discussed, the normalcy or developmental primacy of results from blaming does not help us in determining the function. I propose to determine the function by looking at what the absence of blame tells us about blame’s nature. If we understand what happens when we don’t blame, we might get a better idea about what we think ought happen when we do blame.Footnote 14 In some sense it shows us what is missing, or what goes wrong, when we don’t blame, and thus what it is we need blame to do.
To do this I rely on a distinction made by Jean Hampton (Reference Hampton and Murphy1988). Hampton distinguishes between three different ways we might respond to moral wrongdoing, depending on how we perceive our own self-worth. The first way of responding to wrongdoing that Hampton discusses is an example of a victim with either very good, or too good, self-esteem. If the victim of the wrong has a very firm belief in their value and rank, they might be beyond resentment, and will therefore not blame their perpetrator (Hampton Reference Hampton and Murphy1988: 57). In other words, the victim is not blaming because they are so sure of their own self-worth that the wrongdoing poses no threat to them.
In the second way of responding, one sees the wrongdoing as evidence that one’s value and rank is lower than one previously thought. This will lead the victim to wonder whether she has accorded herself too high a value. Importantly, this does not mean that she thinks the wrongdoing itself has diminished her value, only that it might have revealed her value to be lower than she thought it was (Hampton Reference Hampton and Murphy1988: 50). It is important to note that what is being lowered is not her actual value, but her self-esteem. In a situation like this it will be common for the victim to emotionally protest, and thus blame, the wrongdoer. The blame function as a way of challenging the claim implicit in the wrongdoing, and thus asking whether her perception of her own self-worth is wrong, and the wrongdoer is right in their evaluation of her as someone who can be treated this way. The victim that blames in this way, has some degree of belief in their own self-worth, but worries that the wrongdoer might be right (Hampton Reference Hampton and Murphy1988: 57).
In the third way of responding the way of feeling diminished is more literal. Here, Hampton uses the example of a victim of sexual assault that does not tell anyone what she has suffered because she thinks that this is the kind of behavior women have to endure from men (Reference Hampton and Murphy1988: 49). The sexual assault victim has been led to believe that her value is actually lower than that of a man. The victim is experiencing the wrongdoing as something that actually affects her value by lowering it (Hampton Reference Hampton and Murphy1988: 51).
Hampton’s examples show us something fundamentally important about the reason why we blame. We blame because we need to challenge the meaning inherent in the wrong action we have been subjected to. We need mouse traps to catch mice, and we need the mental state of blame because unrefuted wrongdoing is a threat to our moral status. The genius of Hampton’s examples lies in that they – through showing us what goes wrong in not blaming – manage to explain what is so important with challenging the meaning of a morally wrong action. It also shows us that blame is not just – as argued by so many extant accounts – a way of reporting someone’s lack of adherence to moral norms, or a wish for them to change their ways in the future. We blame because we need some sort of reassurance of our moral worth.Footnote 15
Blame cannot simply be a signal about what norms the victim adhere to, because it is an address that expects, and needs a reply from the wrongdoer. Blame is not a monologue. The dialogue blame starts has clear conditions on its own conclusion. The victim needs for the wrongdoer to acknowledge the wrong and, in Pamela Hieronymi’s words, take back the threatening claim implicit in the act (Reference Hieronymi2001: 548). The claim implicit in the wrongdoing says something about the moral worth of the victim and will remain there as a threat to her self-esteem until it has been acknowledged and taken back. Blame protects the victim’s understanding of their own moral worth.Footnote 16
To bolster this understanding of the function of blame I will quickly survey similar arguments from the literature. Angela Smith writes that: “While it may not in fact ever be communicated to another, blame by its nature has an expressive point and a broadly communicative aim: it expresses protest, and I submit, it implicitly seeks some kind of moral reply (Smith Reference Smith, Coats and Tognazzini2013: 39). To understand better what is being protested through blame, Smith draws on Pamela Hieronymi’s account of blame (Reference Hieronymi2001) and argues that we can see moral wrongdoing as an action that carries meaning about how the wrongdoer evaluates the victim. Another way to put this is that the blameworthy action carries agent-meaning: it is an expression of the agent’s quality of will (McKenna Reference McKenna2016: 247). When performing a morally wrong action the wrongdoer is expressing disregard and is in effect dis-evaluating the victim. The bad action is a statement that communicates how the wrongdoer thinks the victim can be treated (Hieronymi Reference Hieronymi2001: 549). What we are communicating when we are blaming can then be understood as a protest of this statement. In effect we are saying that we are not the kind of person that can be treated this way. The victim protests the meaning implicit in the wrong action, and challenges the wrongdoer to restate the meaning in a way that allows the victim to see themselves as someone who does not deserve to be treated this way.
The important take-away from this section is that when we fail to blame, we are not questioning, or protesting the understanding of our moral worth that is being fronted by the wrongdoing. We can fail to blame because we are too sure of ourselves or too insecure. To successfully protest against the claim implicit in the wrong action, we must communicate, and the communication must have uptake. This shows us what the paradigm-example of blame must be, because the paradigm is the only version of blame that can successfully fulfil the function; that of protesting and standing up for yourself in order to get acknowledgement and reassurance of your moral worth.
5. Benefits of the functional account
In this section I will explain how the proposed version of a functional account can meet the first three data points from the introduction; how it avoids circularity; and how it avoids reverting to conceptual analysis.
The three first data-points can easily be met by this account, as this is one of the main reasons why some philosophers have embraced functional accounts. We don’t need to rely on necessary or sufficient conditions to explain whether an instance of blame in fact falls in under the concept of blame. Emotional, non-emotional, private, and expressed blame can all be accounted for – as long as the instance of blame can be showed to have a residue of the function of blame. The function of blame, I have argued, is to protest and stand up for oneself – or others – against the threat implicit in the wrong action. As we have seen, whether or not an instance of blame actually falls in under the scope of the theory does not depend on how close it is to the paradigm, rather, it depends on whether it has some residue of the function of blame. We can see how non-emotional and private blame has a residue of the function of blame by again considering the case of those not able to blame. The too insecure, or too secure victim of wrongdoing are not even recognizing themselves as victims, or the wrongdoing as something that should be protested. The non-emotional or private blamer, however, is still recognizing the threat, they are just not responding paradigmatically towards it. As such, the blame from these victims will likely not have its function fulfilled. Nonetheless we can recognize it as blame, as opposed to not blame, because the blamer recognizes the threat implicit in the wrong action.
The account I have proposed avoids circularity by first determining the function of blame, and by doing this not by relying on a paradigm-example of blame, but by considering a fundamental element of blame that becomes evident when considering the absence of blame. The consequences of not blaming shows us what it is we want from blame in contemporary social interaction and places blame naturally in our moral discourse. By defining the function first, and then deriving a paradigm-example from this, the account also avoids reverting to conceptual analysis. You don’t have to actually protest the wrong in order to blame. As long as the victim recognizes the wrong action for what it is, the instance has enough of a residue to fall in under the concept of ‘blame’. One might here naturally ask whether relying on the function of blame this way will not admit of false positives. One false positive that might seem relevant to this account, as it is to the more traditional protest account, has been put forth by Shoemaker and Vargas (Reference Shoemaker and Vargas2019). Shoemaker and Vargas use the example of a person quietly protesting execution outside a prison (Reference Shoemaker and Vargas2019: 584). The person is clearly protesting, but it is less clear that they are also blaming. One might say that it is clear that the person is recognizing and protesting the wrong implicit in taking someone else’s life, but that it is not clear that they are also blaming. On my account we can also say, however, that the act of quietly protesting outside the prison clearly has a residue of the function that paradigmatic blame has: the person is protecting our understanding of the prisoner’s moral worth and challenging the statement implicit in an act of execution.
6. Objections: praise and non-moral blame
In this section I will consider some possible further objection to my account. I will do so specifically by looking at a similar, very recent, functional account. Lastly, I will show how my account can meet the fourth, and final, desiderata.
Shoemaker (Reference Shoemaker2024) argues in his latest book that his account is the only one that is able to – in addition to meeting the data points discussed above – explain the symmetry between praise and blame, and account for the nature of non-moral blame. To meet all these data points, Shoemaker modifies his earlier theory – the costly signalling theory (Shoemaker & Vargas Reference Shoemaker and Vargas2019) – by connecting it to the new Norm Maintenance Theory (NMT). In short, Shoemaker’s theory says that we blame to maintain norms. We do this first – to separate it from other norm-maintaining social systems – by delivering a sting in case of blame, or a buzz in case of praise (Shoemaker Reference Shoemaker2024: 27). The sting that comes with an act of blaming gives us an obvious incentive to stick to the norms in the future. Further, because blaming is costly – by blaming we put ourselves in a position where we are in danger of physical and psychological harm – it is a good way for us to signal our commitments to certain norms in a trustworthy way (Shoemaker Reference Shoemaker2024: 31). On Shoemaker’s account then, the two elements of blame – the sting, and the costly signalling – jointly contribute to the function of blame, which is to maintain norms.
Shoemaker assumes that praise and blame should be symmetrical, and thus that an account of blame also should be able to explain the nature of praise. On the account proposed here, this means that it should be able to explain the function of praise and derive the paradigmatic form of praise from this function. First off, I am not convinced that praise and blame ought to be symmetrical. The first asymmetry already manifests itself when we ask what it would look like if we fail to praise. Failing to praise is not necessarily a problem for the one who is failing to praise, it is rather a problem for the praiseworthy. In the case of blame, however, we tend to think that blame is first and foremost important for the victim, and often rather a burden for the blameworthy. There is thus a fundamental asymmetry in the function of the two responses: Blame’s function is mainly to benefit the victim, while praise’s function is mainly to benefit the praised. This asymmetry aside I see no problem with accounting for praise as a way of acknowledging the value or implicit meaning of the praiseworthy action. I do however think that we need more arguments to the fact that blame and praise are symmetrical enough to warrant a shared explanation of their nature.
In relation to non-moral blame, it is controversial whether we can – or should – connect moral blame to blame and praise for non-moral normative failures.Footnote 17 I am inclined to think that many instances of what we might think of as non-moral blame in fact is moral blame if further examined, and that much, if not all, of what is left is not blame, but rather expressions of personal taste. This is not to say that my account wouldn’t be able to explain the nature of non-moral blame, in the few instances it might exist. I assume that non-moral blame would have a function similar to moral-blame and thus be a way for us to stand up against the value-statement implicit in, for example, a ‘bad’ piece of art.
Lastly, my account is able to explain the fourth desiderata from the introduction. When we understand the importance of blame for a victim’s understanding of their own moral worth, the communicative and dialogical element of blame becomes evident. This picture of blame as communicative explains the connection between blame and our other moral responsibility practices. We blame partly because we want the wrongdoer to withdraw the claim implicit in their action, and in this way neutralize or remove the threat to our moral status and self-esteem. Through an explanation, remorse or apology, the wrongdoer can then make blame superfluous, and open up for the possibility of reconciliation, repair and forgiveness. We are not expecting a response when we blame the rain for ruining our wedding day, and we’re not necessarily expecting a response when we are signalling our adherence to a moral norm. But when we are blaming someone in order to protect our moral worth, uptake, and hopefully a response from the wrongdoer, is what we want.
7. Conclusion
In this paper I set out to evaluate the success of the new methods that have been proposed as alternatives to traditional conceptual analysis in the attempt to account for the nature of blame. My aim was to see whether these theories of blame can better account for the four main considerations that have been put forth as important for an account of blame. I concluded that these theories face a problem when it comes to accounting for a plausible paradigm, or function of blame, without becoming circular or reverting back to conceptual analysis. As a response I prosed a new version of the functional account by first establishing the function for blame. I get to the function for blame by looking at what happens when we fail to blame, and conclude that we blame because we need to protect our understanding of our moral worth, or the moral worth of the victim. I have showed how this account can meet all the desiderata mentioned in the introduction without becoming circular and without reverting to conceptual analysis. Lastly, I have answered to some possible objections to the suggested account.Footnote 18
Open access
