different contexts. Through this, they can actually infer, for
instance, a causal relationship between a mutation and a cellular
behavior, given a specific class of contexts. For this reason, some-
times biologists have thought that there are properties that certain
biological entities have regardless of the context. However, much
difficulty in cancer research has arisen at this point. There are
different cases in which, for example, transplanted cells finally lack
the mutated gene, while still retaining the neoplastic phenotype.
In order to better explain these aspects, let us introduce
another notion that can account for the fact that even seemingly
internal properties are in fact relational properties—i.e., the context
determines and constrains the behavior of single biological entities.
This is the notion ofontological dependency. As Wolff [16] rightly
points out, dependency has some advantages with respect to similar
notions such as supervenience and reduction. Unlike superveni-
ence, ontological dependency is anexplanatory relation—i.e., we
explain the existence of an objectxin terms of another object
yinstead of vaguely saying that anytimexoccursyoccurs too.
Unlike reduction, dependence isnot eliminative—i.e., to say that
xdepends onyfor its existence does not eliminatexlike saying that
xisnothing but y. In other words, “[t]o say thatAontologically
depends onBis to say that bothAandBexist, but thatBis in some
sense ontologically and explanatorily prior toA (...)A exists
(at least in part)because Bexists” (p. 618). So the claim we defend
here is that the fact that biological entities only have relational
properties is to be ascribed to the fact that they depend ontolog-
ically on the context they happen to be in, and that their identifica-
tion and the explanation of their persistence through change are
mediated by reciprocal relationships. The illusionary view that some
properties of biological entities are strictly internal is based on
structural similarities between contexts, implying that biological
entities will behave in a similar way across several contexts.4 Evidence from the Biological Sciences
4.1 Intrinsic
Properties
as Relational
Properties
In Systems Biology it is now common to recognize that properties
of a single gene or a single molecule emerge from the properties of
huge networks and from the position of the single gene or molecule
within those huge networks [17]. Palumbo et al. [18, 19] put it
very explicitly: “A gene is defined as ‘essential’ if its deletion has
lethal effects for the organism under a given experimental condi-
tion,” but essentiality, while being a property of the gene, “is an
emergent property of metabolic network wiring.” The authors
name this the “essentiality-by-location” principle. Palumbo et al.
[19] demonstrated that a double mutation involving two enzymes
in yeast, not essential per se, causes death of the organismifthe
double knockout provokes a “lack of alternative path” condition in6 Marta Bertolaso and Emanuele Ratti