The demise of ontology in string theory opens new perspectives on the positions of Quine and Larry Laudan. Laudan stressed the discontinuity of ontological claims throughout the history of scientific theories. String theory’s comment on this observation is very clear: The ontological claim is no appropriate element of highly developed physical theories. External ontological objects are reduced to the status of an approximative concept that only makes sense as long as one does not look too closely into the theory’s mathematical fine-structure. While one may consider the electron to be an object like a table, just smaller, the same verdict on, let’s say, a type IIB superstring is not justifiable. In this light it is evident that an ontological understanding of scientific objects cannot have any realist quality and must always be preliminary. Its specific form naturally depends on the type of approximation. Eventually all ontological claims are bound to evaporate in the complex structures of advanced physics. String theory thus confirms Laudan’s assertion and integrates it into a solid physical background picture.
In a remarkable way string theory awards new topicality to Quine’s notion of underdeterminism. The string theoretical scale-limit to new phenomenology that makes Quine’s concept of a theoretical scheme fits all possible phenomenological data. In a sense string theory moves Quine’s concept from the regime of abstract and shadowy philosophical definitions to the regime of the physically meaningful. Quine’s notion of underdeterminism also remains unaffected by the emerging principle of theoretical uniqueness, which so seriously undermines the position of modest underdeterminism. Since theoretical uniqueness reveals itself in the context of new so far undetected phenomenology, Quine’s purely ontological approach remains safely beyond its grasp. But the best is still to come: The various equivalent superstring theories appear as empirically equivalent but ‘logically incompatible’ theories of the very type implied by Quine’s underdeterminism hypothesis. The different string theories are not theoretically incompatible and unrelated concepts. On the contrary they are merely different representations of one overall theoretical structure. Incompatible are the ontological claims which can be imputed to the various representations. It is only at this level that Quine’s conjecture applies to string theory. And it is only at this level that it can be meaningful at all. Quine is no adherent of external realism and thus can afford a very wide interpretation of the notion ‘ontological object’. For him a world view’s ontology can well comprise oddities like spacetime points or mathematical sets. In this light the duality phenomenon could be taken to imply a shift of ontology away from an external ‘corporal’ regime towards a purely mathematical one.
To put external and mathematical ontologies into the same category blurs the central message the new physical developments have in store for philosophy of science. This message emerges much clearer if formulated within the conceptual framework of scientific realism: An extrapolation of the notion ‘external ontological object’ from the visible to the invisible regime remains possible up to quantum field theory if one wants to have it. It fails fundamentally at the stage of string theory. String theory simply is no theory about invisible external objects.