| | Gravitons
General Relativity (GR) and Quantum Mechanics (QM) were developed
simultaneously to describe phenomena at opposite ends of the scale. GR
describes the interactions of very large masses and interactions over
great distances (where the time required for light to travel is
significant). QM describes the interactions of the very smallest
masses and interactions over very small distances.
In trying to develop a single 'theory of everything' (aka ToE,
analogous to Russell and Whitehead's early 20th century effort to
derive all mathematics from first principles in Principia Mathematica),
it was soon realized that GR and QM are in certain areas
Gravity is simply described in GR as a conserved scalar field
associated with all massive particles that happens to be distributed
throughout geometric space as a 'curvature' (implying a certain type of
variation with distance).
One effort to synthesize GR and QM was to postulate a QM analogue to
the quantized nature of electromagnetic phenomena in the atom. This is
where the idea of 'gravitons' came from. Unfortunately, the requisite
properties of a 'graviton' are such that they are extremely difficult
to observe, and in fact they have never been observed, despite the best
efforts of particle physicists.
The alternative to quantizing GR in this way is adding GR to the
equations of quantum mechanics. Unfortunately, even something as
simple as solving the Scroedinger equation, when attempted in curved
space, exceeds the capacity of our current mathematics.
We know that gravitation exists, and that GR describes it well, and
that QM cannot describe it well. OTOH, We know that there are things
that QM describes superbly well (electron orbitals with unit angular
momenta) that simply fall outside the scope of GR. Both theories are
verifiably correct as far as we have been able to test them. Neither
theory is complete.
What is the answer? My best guess (based on a partially historical
persective) is that we will need to develop a mathematics that is more
general than either QM or GR, and which will 'simplify' to either under
the appropriate limiting conditions.
The situation appears to be similar to that of the photon, which can be
'seen' as either a particle or as a wave. When treated as a
(Lorentzian)four-tensor that satisfies Maxwell's (Unified) Equations in
(Minkowski) space-time, the photon reduces to either a wave or a
particle depending on the operation(s) performed to observe it.
Many physical 'models' - actually mathematical constructs that *mimic*
the operation of 'real-world' objects, processes, and events - suffer
from being too similar to other more ordinary objects. processes, and
events. This similarity is often so strong that the word 'similar' is
replaced in the minds of people with the word 'identical,' so that
people tend to think (for example) that photons *are* waves or that
they *are* particles.
The most honest description is simply a summary of the properties of
something: "we know it has these attributes, and that it interacts with
these specific things in these specific ways."
Short answer to your question: the 'graviton' is simply a hypothetical
particle conceived of in an effort to find a common theory that unifies
both GR and QM. It was developed by particle physicists who were
trying to apply their particle interaction models to gravitation. It
*remains* hypothetical, in spite of the best effots of those who
developed the concept to demonstrate its validity empirically.
"When the only tool you have is a hammer, then every problem begins to
look like a nail." - Abraham Maslow