"We will continue to employ the methodological concept of the model by setting out precisely what its attributes and properties are:
a) It refines certain characteristics of the concept in general, and of conceptual elaboration or conceptualization. By summing up an experimental and practical given, the classic consept turned too much towards the past, and also towards the simple. The model is a more flexible tool, capable of exploring the complex and the random. With it, thought becomes 'propositional' in a new way: programmatic. However, if the 'model' refines the concept, it cannot dispense with it. It presupposes a conceptual elaboration.
b) Like the concept, the model is a scientific abstraction and a level of abstraction. It is always revisable, and cannot be taken either as a reality or entity imminent to the real beneath the appearances of phenomena (the ontological temptation, which structuralism finds hard to avoid), or as a norm or value (the normative temptation). The methodology of models forbids their fetishization.
c) The model is constructed in order to confront 'reality' (experience and practice). It is useful, not least because it helps us to appreciate the gap between itself and the facts, between the abstract and the concrete, between what has been certified and what is still possible. The model is useful: it is a working implement for knowledge. Only the concept has the dignity of knowing.
d) As far as a set of facts is concerned, there can be no question of a single model. If we are to grasp the actual and the possible,we must construct several models. The confrontation between these will be as interesting theoretically as the confrontation between one of them and the concrete element it represents. In this way diversity and discussion during the process take on added value. No one model can be sufficient or pretend to be sufficient by bringing research to a halt. So we are faced with two alternatives: ontology or criticism, dogmatism or empiricism (or pure relativism).
e) The concept of the model also helps to refine the concept of hypothesis. Every model encompasses a hypothesis (in the broadest sense, theoretical or strategic). Every hypothesis concludes by constructing a model, which is the halfway stage between inventing the hypothesis and proving it. So the model assumes the qualities of the hypothesis: provability, creativity. As Politzer said, it should enable us to move from philosophical luxury to the economy of philosophy, by separating the hypothesis out from speculation.
f) Because it must prove its creativity, the model must have an operating or operational character. However, this trait must not be fetishized. The operating techniques linked to a particular model must be examined with care and suspicion. Fetishization of this characteristic, which blows it up out of all proportion, is the feature of a certain well-defined ideology, namely technocratism. The operational model becomes the practical and theoretical property of a bureaucracy and a technocracy. This brings us back to the most disturbing aspect of structuralism. The fetishization of the concept of the 'model' is part of the strategy of the social group of technocrats."
–Henri Lefebvre, Critique of Everyday Life; Volume 2, p. 177.