The normativity of artefacts

https://doi.org/10.1016/j.shpsa.2005.12.006Get rights and content

Abstract

Part of the distinction between artefacts, objects made by humans for particular purposes, and natural objects is that artefacts are subject to normative judgements. A drill, say, can be a good drill or a poor drill, it can function well or correctly or it can malfunction. In this paper I investigate how such judgements fit into the domain of the normative in general and what the grounds for their normativity are. Taking as a starting point a general characterization of normativity proposed by Dancy, I argue how statements such as ‘this is a good drill’ or ‘this drill is malfunctioning’ can be seen to express normative facts, or the content of normative statements. What they say is that a user who has a desire to achieve a particular relevant outcome has a reason to use, or not to use, the artefact in question. Next this analysis is extended to show that not just statements that say that an artefact performs its function well or poorly, but all statements that ascribe a function to an artefact can be seen as expressing a normative fact. On this approach the normativity of artefacts is analyzed in terms of reasons on grounds of practical, and to a lesser extent theoretical, rationality. I close by investigating briefly to what extent reasons on moral grounds are, in the analysis adopted here, involved in the normativity of artefacts.

Introduction

People use artefacts to change the state of the world and sometimes to keep it from changing. Artefact use shares in the broadly normative assessment to which all human activity is liable. Using x to do y may be right or wrong, it may be wise or clever or stupid, it may be skilfully done or clumsily, and so forth. Additionally, the role played by the artefacts in our activities is itself subject to judgements that are broadly normative. An artefact may be more or less fit for its job, may perform its function well or poorly or may malfunction. Although in our use of objects we are not restricted to artefacts, since we can and do use natural objects for all kinds of purposes, the extent to which broadly normative judgements are considered to be in order concerning the performance of objects in general matches the extent to which such objects are ‘incorporated’ into the fabric of human activity. If we choose to use a stone to hammer a pin into the ground, we pick one that ‘makes a good hammer’, but when it cracks after some blows, we do not easily say that our hammer is broken or malfunctioning, as we would if we had been using a ‘proper’ hammer for the job and the handle broke. It seems that, within the context of the human use of objects, the notion of malfunctioning is restricted to artefacts, that is, objects designed and made by human beings for some purpose or other.

The assessments that were presented in the previous paragraph were characterized as ‘broadly normative’. There seems to be a general recognition that our statements judging that something is ‘right’, ‘wrong’, that something ‘ought to be the case’ or ‘ought to be done’, that something or someone is ‘good’ or ‘bad’, ‘recommendable’ or ‘admirable’, and the like, somehow belong together. Such expressions belong to the sphere of human intentional activity. None of these terms applies to the realm of nature in so far as we consider that realm as existing independent of human intentional activity.1

How these various terms hang together, however, is not something on which much of a consensus appears to exist. Usually a division is made between deontic notions (‘right’, ‘wrong’, ‘reason to’, ‘ought to’) and evaluative or axiological notions (‘good’, ‘bad’, ‘evil’). Opinions differ on whether either of these can be considered primary and the other as secondary. Additionally there are conflicting views on how far the value part extends. Some declare instrumental value, being the value of something as a means to an end, not to be a form of value at all,2 while others make instrumental value one of the cornerstones of their categorisation of value concepts.3

In this contribution my aim is to clarify the various forms of broadly normative judgements that are applied to artefacts, and to show how they hang together. I cannot, nor do I need to, evaluate the debate on the precise interrelations between the various normative concepts. I will adopt a characterization of the normative that I think is broad enough to capture both deontic notions such as ‘right’, ‘reason to’ and ‘ought to’ and evaluative notions such as ‘good’ and ‘better’, ‘poor’ and ‘worse’, ‘dysfunctional’ and ‘malfunctional’ that apply to artefact use.

Section snippets

Characterizing normativity

Since there is no accepted general account of normativity, I will adopt a particular one that I find useful for my purpose. It is one sketched very recently by Dancy.4 According to this characterization, the normative concerns the difference that facts about the world make to what to do, believe or desire. What may be termed a normative fact, or, alternatively, what is the content of a normative statement,

Normative judgements of artefact performance

Having articulated, in this way, some of the conceptual apparatus necessary for analyzing normative statements, we can set out to investigate how evaluative judgements, and in particular evaluative judgements regarding artefacts, fit into the realm of the normative.

According to Dancy, evaluative judgements, or facts about values, as his cognitivist perspective has it, are normative but in a vague way. When we say that a particular violin is good, we express the fact that the violin has certain

Additional conditions for the (un)reasonableness of use

The above analysis is incomplete, however. Consider the following case: x is a good car, in that it has features such that p, who needs to do the shopping for the coming week, has a reason to use it to drive to the local supermarket. Does this mean that q similarly has a reason to use the car to drive to her local school? Not if q is twelve years old. In this case she definitely has a reason not to use the car to drive to school. The use of an artefact for the purpose it is designed for usually

Type and token judgements

It should be noted that ‘poor’ and ‘good’ can be applied to artefact types and to artefact tokens. By an artefact token I mean one particular artefact, for example, ‘this knife over here’, ‘my neighbour’s car’, ‘the rifle Lee Harvey Oswald killed John F. Kennedy with’ (if indeed he did). Artefact types come in two varieties, for which I introduce the following two technical terms. By an artefact kind I mean an artefact as defined by its functional role and by the mere fact of being designed to

Criteria for judgements about artefact performance

Why does the presence of certain features f in an artefact x make it the case that a person p has a reason to use x? In the case of instrumental value, it is because the result of the physical process of applying features f, as specified in x’s use plan, is exactly what p hopes to achieve. The fact that a particular artefact is good or poor matches certain criteria that people use to judge the result of using the artefact. If p judges knife x to be better than knife y, it is because p prefers

The inherent normativity of function ascription

Tying together the various results of the previous sections, it becomes apparent that as a consequence of the present analysis function ascription itself is already normative. The analysis is slightly complicated, however, due to the interference of tokens, types and kinds, as the following analysis shows:

x is a working K’ expresses the normative fact that x has features f and that because of these features, if a person p wishes to achieve the result of K-ing, then p has a reason to use x for

Rational and moral grounds

In the previous sections an analysis of normative judgements about artefacts was proposed in so far as these judgements directly relate to the artefacts’ instrumental value. Of the two grounds for reasons, mentioned in the Introduction, instrumental value is linked exclusively to rational grounds. If x is a good K, then x’s features are for any p who wishes to do what a K is for a reason to use x for K-ing on grounds of rationality. In the remaining part of this contribution I will briefly

Conclusion

In the above I have shown that Dancy’s recent characterization of normativity can be used to give a general account of how evaluative statements about artefacts such as ‘x is a good drill’, ‘y is a malfunctioning drill’, and even ‘z is a working drill’, can all be seen as expressing the same type of normative fact. The basic feature of Dancy’s characterization of normative facts is that they express that certain facts are of relevance to what to do, what to believe and what to desire. Without

Acknowledgements

I am grateful to my colleagues participating in the research program The Dual Nature of Technical Artefacts, especially Wybo Houkes, Peter Kroes and Jeroen de Ridder, and to Sven Ove Hansson and to attendants of the Philosophical Reflections on Technical Knowledge conference held in Boxmeer, June 2002, for their comments on earlier versions of this paper.

References (10)

  • W. Houkes et al.

    Design and use as plans: An action-theoretic account

    Design Studies

    (2002)
  • J. Dancy

    Should we pass the buck?.

  • Dancy, J. (Forthcoming). Ethical non-naturalism. In D. Copp (Ed.), The Oxford handbook of ethical theory. Oxford & New...
  • Hansson, S.O. (Forthcoming). Category-specified value statements....
  • Houkes, W. (2006). Knowledge of artefact functions. Studies in History and Philosophy of Science, 37, this issue....
There are more references available in the full text version of this article.

Cited by (0)

View full text