Normative and descriptive statements

  • Posted on: 30 December 2015
  • By: govert

Quite some of the problems that students run in to can be mended by distinguishing more clearly between normative and descriptive statements. Examples of such mistakes are trying to solve a societal problem by merely describing how relevant parts of society are arranged while lacking a specification of the actual solution; or describing a situation and then offering a ‘solution’ to that situation without stating what the problem actually was.

For the sake of explanation, let us first start with rough definitions of normative and descriptive statements.

  • Descriptive statements present an account of how the world is. The word is connected to 'description'.
  • Normative statements present an evaluative account, or an account of how the world should be. The word contains the stem 'norm': something that should be lived up to; or that should be pursued.

In the light of this distinction, some typically descriptive statements are:

  • Michael Jackson died in 2009;
  • Most tree leafs are some shade of green;
  • According to the theory of relativity, the speed of light is independent from the point of reference.

Similarly, some typically normative statements are:

  • Michael Jackson was the greatest musician ever;
  • I love forests of green trees. (Note: this one can be debated; it might just be an empirical account of the fact that I love forests. However, the statement clearly has a performative content, and expresses my value judgment of beautiful forests. Therefore, it qualifies as a normative statement!)
  • Human-driven cars should never go faster than 100 km/h.

If you are not particularly puzzled by the descriptive/normative boundary, you may stop reading here, as the distinction above serves 99% of the needs. However, several different definitions and dichotomies are used in the overwhelming literature of philosophy. I go into that briefly, below.

Is normative the same as moral and ethical?

No it is not. Indeed, moral and ethical claims are generally normative; but they do not exhaust all possibilities. It is best to see moral and ethical claims as a subset of all normative claims. A circle should be drawn using a compass – this is by its semantic form clearly a normative phrase, and if you are told so by your maths teacher, you will understand that it is a normative statement that you better comply with. Yet, there is nothing particularly moral or ethical about it. The claim ‘Beethoven was the greatest composer of all times’ is clearly a normative claim. It evaluates the artistic genius of Beethoven, and provides a measure long which we can judge other composers. However, there is little immoral about ‘being not such a great composer as Beethoven’. It is rather a matter of taste, or perhaps a judgment by someone who is particularly authoritative because of some special expertise. The same holds for ‘the weather is fine’, or ‘it is ridiculous to spend € 1000 for a pair of jeans’.

The specific subset of normative claims that forms the set of moral claims is generally agreed upon to be characterized by the following properties.

  • Moral claims concern fundamental rights or goods, such as the protection of life, freedom, bodily integrity and well-being; and
  • Moral claims have a strong pretention to universalization. They are typically such that we do not only believe that we ourselves should accept them, but that it is also necessary that others do so (which is not to say that they actually do, nor that we have the right or means to force them to).

For all practical purposes, moral claims and ethical claims are one and the same. However, sometimes the distinction is made that morality is the set of norms that we adhere to unproblematically, while ethics is the activity of reflecting on morality and its problems. This entails that moral claims concern the uncontroversial claims that we need not discuss (or at least at some point can leave undiscussed), while ethical claims are controversial and typically subject of discussion.

Normativity and its antonyms

Usually, it is not the most convincing argumentative strategy to explain something by examples of what it is not. If you are asked what a football is, do not start with explaining how it is in any way not a tennis ball. However, the problem with normativity is that it actually has several antonyms (concepts that oppose it), and it is helpful to explore them.

Primarily, normative is used as being opposed to descriptive as explained above. The demarcation between the two is on the difference between describing an actual state and a desired state. I will henceforth call this ‘normative-as-prescriptive’.

However, sometimes, the demarcation is thought to be on a different line. Normative is then used as opposed to empirical, and the demarcation is on the question whether or not the claim is for its validity dependent on knowledge of ‘the world out there’. In this sense, mathematical knowledge is highly normative. In fact, mathematicians are the champions of finding knowledge that is overtly independent of knowledge about the world. A square has four sides and its area equals the length of one side multiplied with itself. Of course this has a meaningful correlate in the actual world, but hard-core mathematicians do not care very much for real-world squares. The knowledge precedes empirical observation, hence my further use of ‘normative-as-apriori’.

Note however, that mathematical knowledge is not at all normative in the first sense, normative-as-prescriptive. In fact, mathematical knowledge is in that sense descriptive: it consists of statements about how things are (be it often in an imaginative space, not the real world). There is nothing particularly beautiful, desirable or morally compelling about the claim that a parabola has exactly one extreme.

It is in the meaning of normative-as-apriori, that theories, in virtually any discipline, can be normative: whenever they precede empirical investigation, instead of being its consequence. Vast parts of cosmology are normative: it contains mathematical constructs that precede empirical investigation. Surely they aim at testability and in the end it is hoped that observations confirm or disprove them, but until then, they are normative (not moral, though). But also much of social theory and philosophy is normative in this sense, even when it only makes descriptive claims. The point is that even when those disciplines make descriptive claims, they often depend on constructs (“complex theoretical concepts”) that do not have a clear observable correlate in the real world. Of course, such constructs are likely normative-as-prescriptive as well, but that need not be their explicit purpose.

Note that both sorts of normativity tend to go together well, though mathematical knowledge epitomises the exception.

Against ‘normative-as-prescriptive’, the word descriptive is generally considered synonymous with 'empirical' and 'positive'. In this sense, normative is also used as opposed to ‘sociological’ by Jürgen Habermas (Between Facts and Norms, Polity Press, 1996, p. 549), which of course refers to a particular realm of the empirical.


One problem concerning normativity is when the normativity is hidden; so-called crypto-normativity. You will easily find statements that are by its form descriptive, i.e. they refer to a situation or state of things, while it is at the same time crystal clear that the speaker intends to convey a judgment or opinion or other intervention. ‘There is a stain on your shirt’ usually conveys the normative content of disapproval, especially if a mother tells it to her child. Sometimes the normativity is only in the tone of voice. With the explanation above, more subtle and ambiguous examples can be unmasked.