You should not think of $\mathbf{F}=m\mathbf{a}$ as defining force. You should think of it as telling you what the acceleration will be under the influence of a known force that depends on position, time, and velocity:
$$m\frac{d^2\mathbf{r}}{dt^2}=\mathbf{F}\left(\mathbf{r},t,\frac{d\mathbf{r}}{dt}\right)$$
It is the equation of motion, a second-order differential equation to be solved for the position $\mathbf{r}(t)$, when one knows the proper formula for the force.
If it were just a definition, it could have no predictive value. As an equation of motion expressing the dynamical evolution of a system, it predicts the future.
The causal relationship is force causes acceleration, not acceleration causes force.
Various kinds of interactions have specific formulas for the force as a function of position, time, and velocity. For example, for two point particles under Newtonian gravity, the force on one depends only on its position relative to the other, and not on time or velocity:
$$\mathbf{F}\left(\mathbf{r},t,\frac{d\mathbf{r}}{dt}\right)=\frac{GMm\,(\mathbf{r}-\mathbf{R})}{|\mathbf{r}-\mathbf{R}|^3}.$$
For a charged particle in an electromagnetic field, the force can depend on position, time, and velocity:
$$\mathbf{F}\left(\mathbf{r},t,\frac{d\mathbf{r}}{dt}\right)=q\left[\mathbf{E}(\mathbf{r},t)+\frac{1}{c}\frac{d\mathbf{r}}{dt}\times\mathbf{B}(\mathbf{r},t)\right].$$
These are the kinds of equations that define specific forces.