A deflationary account of information in biology

An oft-repeated claim is that there is information in some biological entity or process, most especially in genes (Downes 2005). Some of these claims derive from the Central Dogma, population genetics, and the neo-Darwinian program. Others derive from attacks upon evolution, in an attempt to show that “information cannot be created” by natural selection. In this paper I will try to show that the term “information” is a homonym for a range of distinct notions, and that these notions are either of concrete properties, in which case they are little more than a periphrasis for correlation and causation, or of abstract properties, in which case they are observer-dependent. In short, if information is in the concrete world, it is causality. If it is abstract, it is in the head