Let's begin by reviewing some grammar. 'Walking' is the present participle of the infinitive 'to walk.' Present participles are formed by adding -ing to the verb stem, in our example, walk. Participles can be used either nominally or adjectivally. A participle used nominally is called a gerund. A gerund is a verbal noun that shares some of the features of a verb and some of the features of a noun. Examples:
Walking is good exercise.
Sally enjoys walking.
Tom prefers running over walking.
As the examples show, gerunds can occur both in subject and in object position.
Participles can also be used adjectivally as in the following examples:
The boy waving the flag is Jack's brother.
Sally is walking.
The man walking is my neighbour.
The man standing is my neighbor Bob; the man sitting is his son Billy Bob.
The Muslim terrorist cut the throat of the praying journalist.
Now what about the dreaded fused participles against which Fowler fulminates? In the following example-pairs the second item features a fused participle:
She likes my singing.
She likes me singing.
John's whistling awoke her.
John whistling awoke her.
Sally hates Tom's cursing.
Sally hates Tom cursing.
If you have a good ear for English, you will intuitively reject the second item in these pairs. They really should grate against your linguistic sensibility even if you don't know what it means to say that gerunds take the possessive. That is, a word immediately preceding a gerund must be in the possessive case. A fused participle, then, is a participle used as a noun preceded by a modifier, whether a noun or a pronoun, that is not in the possessive.
Fused participles, most of them anyway, are examples of bad grammar. But why exactly? Is it just a matter of non-standard, 'uneducated,' usage? 'I ain't hungry' is bad English but it is not illogical. Fused participles are not just bad usage, but logically bad inasmuch as they elide a distinction, confusing what is different.
This emerges when we note that the members of each pair are not interchangeable salva significatione. It could be that she likes my singing, but she doesn't like me. And if she doesn't like me, then she doesn't like me singing or doing anything else.
In the second example, it could be that the first sentence is false but the second true. It could be that John, who was whistling, awoke her, but it was not his whistling that awoke her, but his thrashing around in bed.
The third example is like the first. It could be that Sally hates the sin, not the sinner. She hates Tom's cursing but she loves Tom, who is cursing.
Is every use of a fused particular avoidable? This sentence sports a fused participle:
The probability of that happening is near zero.
The fused participle is avoided by rewriting the sentence as
The probability of that event's happening is near zero.
But is the original sentence ungrammatical without the rewriting? Technically, yes. One should write
The probability of that's happening is near zero
although that is perhaps not as idiomatic as the original. In any case, one would have to be quite the grammar nazi to spill red ink over this one.
According to Panayot Butchvarov, "Fused participles are bad logic, not just bad usage." ("Facts" in Cumpa, ed., Studies in the Ontology of Reinhardt Grossmann, Ontos Verlag, 2010, p. 87.) In Skepticism in Ethics, Butch claims that a fused participle such as 'John flipping the switch' is as "grammatically corrupt" as 'I flipping the switch.' (Indiana UP, 1989, p. 14.)
I think Butch goes too far here. Consider the sentence I wrote above:
And if she doesn't like me, then she doesn't like me singing or doing anything else.
I don't agree that this sentence is grammatically corrupt. It strikes me as grammatically acceptable, fused participle and all. It expresses a clear thought, one that is different from the thought expressed by
And if she doesn't like me, then she doesn't like my singing or my doing anything else.
The first is true, the second false.
So we ought not say that every use of a fused participle is grammatically corrupt. We ought to say that fused participles are to be avoided because they elide the distinctions illustrated by the above three contrasts. The trouble with 'I hate my daughter flunking the exam' is not that it is ungrammatical but that it fails to express the thought that the speaker (in the vast majority of contexts) has in mind, namely, that the object of hatred is the flunking not the daughter.
What does this have to do with ontology?
Some of us maintain that a contingent sentence such as 'John is whistling' cannot just be true: it has need of an ontological ground of its being true. In other words, it has need of a truth-maker. Facts are popular candidates for the office of truth-maker. Thus some of us want to say that the truth-maker of 'John is whistling' is the fact of John's whistling. Butchvarov, however, rejects facts. One of his arguments is that we have no way of referring to them. Sentence are not names, and so cannot be used to refer to facts.
But 'John's whistling' fares no better. It stands for a whistling which is an action or doing. It does not stand for a fact. So some use fused participles to refer to facts. Thus, the fact of John whistling. Butch scotches this idea on the ground that fused participles are "bad logic" and "grammatically corrupt."
I don't find Butchvarov's argument compelling. As I argued above, there are sentences featuring fused participles that are perfectly grammatical and express definite thoughts. My example, again, is 'If she doesn't like me, then she doesn't like me singing or doing anything else.' So I don't see why 'John whistling' cannot be used as a name of the fact that is the truth-maker of 'John is whistling.'