No, "troll" because you're trolling. Make an argument that isn't a textbook logical fallacy.
>About as provable as the opposite.
Are you arguing semantics? Absolute proof of anything is impossible? The thing to be proven is insufficiently specified, e.g. whether limited to software patents or to historically vs. in today's economy? That sort of thing? That's extremely pedantic, but I suppose it's technically correct. (The best kind of correct, naturally.)
So let's say not proof, just good reasoning supporting the position. Here, I'll start.
In the software industry where every individual with a PC and a compiler is a market participant, there are thousands of times more players than there are in most other patent-intensive industries, which vastly increases the cost of the patent system because the cost of clearing and licensing each invention against each patent is O(n^2) in the number of inventions. This number is so large in a market with tens or hundreds of thousands of participants that actually doing the clearances and licensing becomes an impossibility in practice due solely to transaction costs, i.e. the cost of each and every inventor tracking down and negotiating with each and every other inventor holding a potentially relevant software patent.
Moreover, as a result of the practical impossibility of actually doing that many clearances, it creates an environment in which everyone is potentially infringing everything. In such an environment, the patent system is inherently not providing the expected incentive to inventors, because there are too many infringers infringing too many patents to even so much as evaluate all of the possible infringement, much less negotiate licenses or litigate it, resulting in no license fees (or monopoly rents through the ability to exclude) from the large majority of infringers.
In addition to that, because patents are legally required to be licensed but software cannot rigorously and thoroughly be determined to be non-infringing in any cost-effective manner, an enormous amount of uncertainty about what patents a product may infringe overhangs the prospective market participants, which reduces the amount of funding available for new R&D because prospective investors will only provide funding if the expected returns exceed the risk created by software patents. Worse, because of the enormous amount of de jure infringement which is de facto business as usual, patent trolls with submarine patents can do an enormous amount of damage to anyone who experiences market success, and there is nothing the victim can do to prevent it because of the aforementioned practical impossibility of doing all the necessary patent clearances. This is worse than a normal risk because it occurs exactly to those investments which would otherwise have succeeded, providing an even more powerful disincentive to invest in new product development. The associated litigation creates a further deadweight loss on the economy because not only are those resources not being allocated to research, they're actively being misallocated to wasteful litigation and transaction costs.
On top of all of that, the reverse can occur and results in a potentially even larger harm: Rather than a successful company being sued by a non-practicing troll, the successful company can sue anyone who actually makes a product, because they hold enough patents that any nontrivial software infringes some of them. Between large companies this power cancels out, but when the relationship is asymmetric it becomes a club for incumbents to use against newcomers. Anyone who enters an established market segment is subject to suit by incumbents, which reduces the incentive for non-incumbents to innovate, because the cost of the inevitable litigation has to be factored in. Where this effect is sufficiently strong, it can destroy innovation entirely in specific submarkets, because no newcomers enter for fear of incumbents with patents and no incumbents innovate for lack of competition from newcomers.
So there you go. Let's hear your rebuttal.