Against Raz' Notion of Second Order Reasons

Table of  Contents  

1.  The Objective  1

2.  Raz Account of Second Order Reasons  1

2.1.  Formalizing Reasons and the Whistleblowing Example  1

2.2.  Where Raz Faces Problems  3

3.  Second Order Reasons as First Order Reasons  4

3.1.  The Counterproposal  4

3.2.  Investing or not Investing that is the Question: Ann s Example  4

3.3.  The Example of the Complying Soldier and the Noncompliant Subordinate  6

4.  Summary and Conclusion  7

  second order reason SOR  

Christoph Siemroth Seminar on Normativity

In the following essay I shall try to show that Joseph Raz' descriptions of second order reasons and exclusionary reasons are not the most plausible way to describe deliberation and decision, because it involves some strange consequences. Though it may be true that accepting his account of second order reasons gives a comfortable order to some seemingly complicated cases, it is not true that they provide the most consistent solution with our intuitions. My suggestion will be to reduce Raz' sec- ond order reasons to first order reasons, which suits our belief on how decisions are made better. Since Raz' intention merely seems to be to show that people act on second order reasons, and this claim is mainly supported by employing some examples, I also shall try to investigate these exam- ples in a different light. After that, it should be clear that his second order reasons are not necessary to explain human behaviour as he claims. In order to narrow down the examination, I will only focus on his descriptions in the first chapter. All references to Raz are meant for:

Raz, Joseph (1975): Practical Reason and Norms, Oxford University Press, 1999.

2. Raz' Account of Second Order Reasons

This chapter serves to repeat and formalize Raz' descriptions. The example is based on Raz' story of Colin (cp. p. 39). Assume the company that hired you produces toxic food and you have the option to make this public in order to protect the consumers (W = whistleblow, act). On the other hand, this leads to an umcomfortable life for your family for the next years, due to press invasion, no employer will hire you again etc. Also, you once made a promise never to even consider options that are contrary to your family's interests. Following Raz, you have an exclusionary reason that sets aside a decision based on the merits of the case, because your family's well-being is at stake. For this to be an exclusionary reason, it must look like that (note that the SOR does not directly take side for a certain option, but rejects you should make a decision on the merits at all):

In this representation, if you were to ignore the SOR, then your choice would be to do W, since the protection of lives is a conclusive (i.e. deciding) reason.

1

Christoph Siemroth Seminar on Normativity

Let us introduce a measure of the strength of reasons, 1 which is a cardinal measure, 2 and call it 'force'. The higher the force WKH VWURQJHUWKHDVVRFLDWHG UHDVRQ í LW LV UDWLRQDO WR GHFLGH RQ WKH option which has the highest accumulated force. In order to dodge objections that such a measure does not exist or is too narrow, let us further assume it is merely a heuristical device to illustrate the strength of bundles of reasons. With example-values, it looks as follows:

Now let us consider the SOR. Even though it requires us not to decide on the merits of the case, by choosing to comply with the SOR, we inevitably also choose an outcome to the decision on the problem at hand, namely, we choose not do to W. This is not because the SOR overrides the FOR on the merits, but because by choosing SOR we are forfeiting the option to do W, and all of that is done (according to Raz) not considering the merits of the case. If we accept to comply with SOR, we are nonetheless making a choice.

The structure of the ordering of the reasons is what John Rawls calls a 'lexical order', which

PHDQV WKDW í DSSOLHG WR WKLV FDVH í WKH RUGHU ZLWK KLJKHU SULRULW\ KHUH WKH KLJKHU RUGHU UHDVRQ

decides alone on the issue at hand, ignoring the lower order reason. Only if there is an indifference on the higher order level, i.e. in SOR, then the lower order has influence on the decision. This is, in my view, a fair characterization of what Raz understands as SORs and how they behave. In his words, there is a "general principle of practical reasoning which determines that exclusionary rea- sons [=SORs] always prevail, when in conflict with first order reasons" (cp. p. 40). Basically, a lexical order means no however huge increase in force for FORs can compensate for a loss in force for SORs. Now we can also assign force-values to SORs, since they still may override each other. The representation, then, looks like this:

1 Cp. p. 25-7. He defines strength only in relative terms, whereas we need an absolute measure to compare bundles of

reasons with other bundles, hence the introduction of the measure.

2 That means differences have meaning, can be compared (such as 5í3=8í6) and thus single values can be added.

2