The issue with saying this is we should probably have many words for different types of depression.
Probably not only split on this one variate (externally or internally triggered), but it means that certain types of depression are usually triggered externally and certain types not externally.
It's like saying that "in some cases, a broken leg is not triggered by some external influence". Sure, you could develop some bone density issue and break a leg when walking normally, so we tend to distinguish treatment of these cases--we may recommend a more permanent solution for someone who is prone to breaking their leg.
How can we even falsify your statement? To do so would require self-reports or some sort of real-time 24/7 measurement of the brain, the first doesn't really falsify anything, and the second is not very viable at the present.
Self-reports can easily lie. They also may simply not remember, or be in denial.
