Thursday, October 7, 2010

The Dialectical Necessity of Morality: Methodology and Terminology

This post is part of my series on Deryck Beyleveld's book The Dialectical Necessity of Morality. In the book, Beyleveld tries to argue in favour of Alan Gewirth's Principle of Generic Consistency (PGC). For an index, see here.

As this is the first truly substantive entry in the series, I will present a brief sketch of the argument to the PGC. I will follow this with a discussion of the methodology underlying the argument. I will close by running through some of the terminology that will be essential for understanding the argument.

1. The Argument in Brief
As stated in the introduction, the PGC is (allegedly) the supreme moral principle that must be accepted by all potentially purposive agents (PPAs). The following is a quick sketch of the argument to the PGC:

  • (1) To be a PPA is to claim: I do X voluntarily for some purpose E (premise).
  • (2) E is good (from 1).
  • (3) Freedom and well-being are generically necessary conditions of agency, i.e. they are necessary no matter what purposes I may wish to pursue (premise).
  • (4) My freedom and well-being are necessary goods (from 2 and 3).
  • (5) I (even if no one else) have a claim right to my freedom and well-being (from 4).
  • (9) Other PPAs (PPAOs) have the same claim right (argument for this is omitted).
  • (13) I must accept that every PPA has a claim right to its freedom and well-being.

As should be apparent from the numbering, several critical portions of the argument are left out of this sketch. Still, enough is presented to give you a feel for what is trying to be proved. Beyleveld (and Gewirth) are trying to show that anyone who thinks that they are a PPA must, on pain of inconsistency, accept that all PPAs have claim rights to their freedom and well-being. This is a very strong claim.

2. Methodology
Beyleveld adopts a particular style of argument which he terms dialectic. He contrasts this with an assertoric method. A dialectic argument is something which appeals directly to the subjectivity of the agent and deduces conclusions from premises to which an agent would (subjectively) be inclined to agree. An assertoric argument would deduce its conclusions from statements about the (objective) properties of agents.

Furthermore, the argument is dialectically necessary, not merely contingent. This is because the premises are ones to which any PPA, in virtue of being a PPA, must agree.

There is an important point to be made about the methodology underlying the argument, particularly if one wishes to assess its metaethical pretensions. Since this is exactly what I wish to assess, I am going to make it, however, it should be noted that, as far as I am aware, neither Beyleveld nor Gewirth see fit to do the same.

The key questions in metaethics are, roughly, do moral terms like "good", "bad", "right", and "wrong" have truth values? Do they refer to actual states of affairs? Are those states of affairs mind-dependent or mind-independent? How can they be known?

If one rejects non-cognitivist positions (i.e. positions maintaining that moral terms do not have truth values) then one must accept that terms such as good, bad, right and wrong can be correctly applied to certain states of affairs. This is to embrace the existence of some identity-relationship between a term such as "good" and an actual state of affairs.

The methodological question is: how can this identity-relationship be established? There are two ways in which this can be done.

First, if the term has some universally agreed-upon referent, then one can investigate the referent and establish the relationship between it and the term. The classic example of this method in action comes from the identity-relationship between "water" and "H2O". Because the term water had a universally agreed-upon referent (the colourless, odourless, liquid that everyone drinks) scientists could work out its chemical structure and then establish the equivalency between H2O and water.

Second, if the term has some widely agreed-upon meaning, then one can derive a set of platitudes about the term and see which state of affairs best instantiates those platitudes. This method is very popular in metaethics where there is considerable disagreement about the referents of moral terms, but some reasonable agreement about the meaning of those terms (or what would need to be the case for them to successfully refer). Indeed, the method is used by Michael Smith in his book The Moral Problem, by Robert M. Adams in his attempt to develop a modified divine command theory, and, in a negative way, by Richard Joyce in his argument in favour of error theory.

It should come as no surprise then to learn that the second method is (implicitly) adopted in Beyleveld's work. He begins with the following four platitudes about moral "oughts" and criteria of practical reasonableness:

  • Universality: "X ought to be done" applies, at least in part, to every agent.
  • Other-regardingness: A moral criterion of practical reasonableness will take favourable account of the interests of agents other than, or in addition to, myself.
  • Conative Independence: The prescription "X ought to be done" holds irrespective of the agent's occurrent wishes and desires.
  • Over-ridingness: A moral criterion of practical reasonableness will take precedence over all other criteria of practical reasonableness. 

Having singled-out these four moral platitudes, the success or failure of the argument to the PGC is easily measured: it will succeed if the PGC satisfies those four platitudes; it will fail if it does not. Consequently, it is difficult to understate the importance of these four platitudes to Beyleveld's (and Gewirth's) enterprise.

One final point about methodology and we will be done. In addition to the four platitudes, there are three central questions in moral philosophy that it is hoped the argument to the PGC can answer. They are:

  • The Authoritative Question: Why should one be moral?
  • The Distributive Question: Whose interests, other than one's own, should be favourably considered when acting?
  • The Substantive Question: Which interests should be taken into consideration?

3. Terminology
One of the more annoying things about the book is the terminology (particularly the acronyms) with which it is overloaded. Because of this, I thought it might be useful to provide some simple explanations of the terminology used throughout the book. I'll add to this as I go along and I will repeat the explanations whenever I introduce a term in future entries.

  • Potentially Purposive Agents (PPAs): These are the agents to whom the argument is directed. A PPA is someone who voluntarily acts so as to achieve purposes or goals. Also "Other PPAs" (PPAOs)
  • Subjective Viewpoint on Practical Reasonableness (SPR): A PPA's personal view, theory or criterion by which it assesses which purposes are permissible/impermissible for it to pursue. Three SPRs are highlighted: (i) the Adeonticist-SPR, according to which there are no impermissible purposes; (ii) the Deontic Amoralist-SPR, according to which there are some restrictions on purposes but none that involve duties to others; and (iii) the Deontic Moralist-SPR, according to which there are other-regarding restrictions on purposes.
  • Claim Rights: The object of the argument is to prove that a PPA must accept that all PPAOs have claim rights to freedom and well-being. A claim right is something that comes with a correlative duty which is supposed to apply to others. A negative claim right comes with a correlative duty to not interfere with the person making the claim. A positive claim right comes with a correlative duty to provide assistance to the person making the claim. A claim right is to be contrasted with a weak right or a "mere liberty" which is a permission to do something with no correlative duty.
  • Generically Necessary Requirements of Agency: These are things that all PPAs, irrespective of purposes they try to fulfil, will require. 
  • Freedom: One of the two (allegedly) necessary requirements of agency. The specific focus is on dispositional freedom, which refers to the general capacity or ability to control one's behaviour by unforced choice. This is to be contrasted with occurrent freedom which refers to the unforced operation of that capacity at a particular moment. It is argued that dispositional freedom is necessary to have any purposivity at all, but that occurrent freedom can be waived in order to achieve specific purposes.
  • Well-being: The second of the (allegedly) necessary requirements of agency. Well-being is said to have three levels. Basic well-being covers the proximate necessary preconditions of performing an action, such as physical fitness and integrity, and mental equilibrium and confidence. Non-subtractive well-being covers whatever the agent needs to maintain what it already has that is good. Additive well-being covers whatever the PPA needs to increase its existing level of purpose-fulfillment. Basic well-being is an absolute necessity for a PPA, nonsubtractive and additive well-being are important for general success as a PPA. One can imagine a hierarchy of well-being with basic at the top and additive at the bottom.

That list barely scratches the surface but it does cover the more important stuff. 

In the next post we will begin to work our way through the stages in the argument to the PGC.


  1. An interesting direction to be sure.

    The main failure of the Kantian approach, I believe, is in establishing a priori anything like (9). Reason may necessitate the pursuit of certain goods, such as truth or well-being, by the individual. But I fail to see how Reason alone can motivate him to respect the right of others to pursue them.

    I have come to the conclusion that metaethics is a mess. My solution is to be practical - I define ethics, arbitrarily, as essentially a body of knowledge that aids normative people fulfill rationally decide their actions. I bring this up because this leads me to deny both the metaethical methodologies you raised. They both were attempts to reach some consensus and capture an existing "real" meaning, whereas I think there are numerous "correct" meanings - the one I use is just one of them; although it is the only one a normative person should use to guide his actions.


  2. When I got around to it, your point about (9) was going to be my major initial objection to this argument.

    Actually, I thought about it differently. My initial feeling was that I could accept that other PPAs would have (from their perspective) rights to freedom and well-being, I just couldn't see how that would commit me to actually respecting their rights in my actions.

    It seems, on closer inspection, that the idea of "claim rights" is what allows Beyleveld to reach such a conclusion. Recognising the existence of a claim right necessarily forces one to accept the existence of a correlative duty. So if you think others have claim rights, you have a duty to respect them.

    I'll be interested to see if Beyleveld can convince me that I have to accept the claim rights of others. His book is loooong, and not an easy read, but nothing so far convinces me that he can.

  3. Yes, we are thinking along the same lines. Do tell if Beyleveld manages to do the impossible - but I'm not holding my breath :)