tom.stoer said:
The Fock space is spanned by states created via creation operators.
Yes, that is one choice of basis, and the one that is almost always chosen. This is sensible, since this method takes advantage of the spacetime translation symmetry for purposes of mathematical simplification and elegance.
It is, however, not necessary to use this momentum eigenstate basis. In principle, for ordinary QFT, one could use a position eigenstate basis where the creation operator generates a point state, without a definite particle number. It would not look as nice since one would not take advantage of the nice translation symmetry of spacetime.
It is evident from the path-integral formulation of an ordinary QFT that initial and final states in an amplitude can be chosen as point states.
Comparing the Fock space of an ordinary QFT and at the Fock space of a quantized string there is not so much difference.
My comment was regarding second quantized string field theory,
not the string worldsheet field theory. As I understand it, creation operators in a string field theory will be completely different. One such creation operator would be accompanied by not only a few quantum numbers (i.e. momentum + spin), but instead a centre of mass momentum together with an infinity of Fourier coefficients to completely describe a general excitation of e.g. a closed string. Much more data is needed to create a one-string state from the vacuum in string field theory, than to create a basic point-state or one-particle momentum eigenstate from the vacuum in ordinary QFT.
Ref:
http://en.wikipedia.org/wiki/String_field_theory
There is no position eigenstate basis that will span the Fock space in the string field theory. This is evident, since this is a quantum theory of closed curves. It is evident from looking at the path-integral formulation, where initial and final states of a quantum amplitude must be chosen as closed curves.
Why do you say "one-particle point states (corresponding to Dirac delta distributions).", why not simply "one-particle states"?
It was a mistake on my part. The state I spoke of is localized to a point, but doesn't have a definite particle number. Products of such states can span the Fock space of ordinary QFT, but not in string field theory.
At least that was my understanding of all this.