Agreed (although there are multiple notions of dimensionality at play here). The trajectory of a thrown javelin can be regarded as one-dimensional for many purposes, but if we look closely enough we find that the javelin itself has width. Even if we neglect its width, it has length; and as its velocity is generally not exactly parallel to its axis, it will sweep out a two-dimensional surface as it flies.
To say, as GP does, that the trajectory of a thrown javelin is "inherently ... one-dimensional" is to regard the simplest approximation as more fundamental than the physical reality -- i.e. confusing the map for the territory.
In my book, the reason why one cannot throw a javelin wide is much simpler: "wide" is not an adverb that can meaningfully modify the verb "throw".