A judicious use of commas helps to avoid misinterpretation. I'd also hazard to note, at the risk of sounding like a get-off-my-lawn prescriptive grammarian, that many garden path sentences achieve their ambiguity via hanging prepositions, e.g.:
>The cotton clothing is made of grows in Mississippi.
Yet the most general pattern seems to be a simple matter of loading too many words into the subject clause [1], e.g.:
>The car driven past the barn crashed.
We don't reach the verb until the last word, which not only opens the door to ambiguity but also produces flaccid prose. You will produce tighter, sharper sentences by front-loading the predicate verb.
[1] - edited from "too many clauses" as per telemachos's astute reply.
English is a Germanic language. (Yes, there's other stuff mixed in, but the root is Germanic.) There is a natural tendency for the main verb to go in second position. However, second position does not necessarily mean second (full stop) because of phrase groupings. Your own sentences give good examples. I've put parens to group things and second position is in italics:
> (A judicious use of commas) helps ...
> (the most general pattern) seems...
(I think that the initial adverbial 'Yet' doesn't count for purposes of verb position, but I may be wrong about that. It's been a bit for me...)
The thing is that the sentence you quote as a bad example (and it is bad, and should be edited) feels natural, because it follows that rule:
> (The car driven past the barn) crashed.
There aren't multiple clauses at the front. There is a single (overpacked) noun phrase.
I'm not excusing bad writing, but the patterns that underlie your own language can trip you up in surprising ways.
You're absolutely right, of course. By "front-load" I don't mean literally the second word; rather, I mean moving the predicate verb as close to the front of the sentence as clarity and flow allow.
"Whenever the literary German dives into a sentence, that is the last you are going to see of him till he emerges on the other side of his Atlantic with his verb in his mouth."
"In a similar vein, Martin Gardner offered the example: Wouldn't the sentence "I want to put a hyphen between the words Fish and And and And and Chips in my Fish-And-Chips sign" have been clearer if quotation marks had been placed before Fish, and between Fish and and, and and and And, and And and and, and and and And, and And and and, and and and Chips, as well as after Chips?"
Garden path: "The horse raced past the barn fell."
The whole meaning is changed by adding one word to the end of the sentence, and you have to read the whole thing again to get it right. Makes hell for software trying to parse the sentence.
Paraprosdokian: "It's too bad that whole families have to be torn apart by something as simple as wild dogs."
The whole meaning is changed (and made funny!) by an unexpected series of words at the end. I love the examples on the wikipedia page.
What's interesting is that by looking at how many nested levels are still understandable, we can also make a guess about the stack size of our brain. (short term memory (?))
In fact, at present four of the 10 front-page Google results for that query were written by me. (Two copies of the one in this thread, plus two copies of something on the C2 wiki. I can't be 100% sure that I wrote the latter, but I'm pretty sure it was me.)
I wonder about the same phenomenon but on higher levels of organization, e.g. concepts in an explanation. Are there valid ways to explain some concept C such that the interpreter is led down a garden path.
More important would be the existence of exclusively garden path concepts, call it the set EGPC. EGPC is the set of concepts which can only be arrived at via explanations or chains of concepts and reasoning that include at least one misleading garden path. One goal of clear teaching would seem to be reducing the number of conceptual garden paths that students encounter; so knowing that some subject necessitates them would be useful.
I'm noticing this while reading Under the Dome. There are many sentences where I just stop at the last couple words because the sentence doesn't make sense at that point. Quite a few times I thought they were typos. When I analyze the sentence though I find that I needed to pause in another spot or use a different meaning of a word. Haven't run into this often in other books.
I didn't run into it much through the next 800 pages of the dome. I'm almost half way through Catch-22 and it is essentially built on this type of sentence.
I've just had a similar conversation with my Japanese wife, and she pointed out that with many of the words that can be interpreted as nouns or verbs by fluent speakers, she only knows one interpretation, so she misses the 'garden' path that a fluent speaker would take. I suspect that it also applies to transitive vs. intransitive verbs.
So to contrive your example, a fluent speaker would know both the transitive and intransitive uses of "returned" and, preferring the intransitive interpretation ("I returned home"), would reach your conclusion. But if a non-fluent speaker only knew -- or was stronger with -- the transitive version ("I returned the book"), they would arrive at the correct interpretation.
I hope that makes sense, it was a lot harder to explain than I anticipated!
I'm a native English speaker, living in a foreign country, so I spend most of my day working with a language I'm not completely fluent in, and I find I do the opposite-- I'm trying to put the meaning together word by word as the sentence unfolds, which makes reading things with long, Proustian sentences very difficult.
I did a presentation a few years ago on some studies about using prosody to resolve sentence ambiguities. The improvement in listener comprehension was measurable and dramatic. The easy conclusion to draw is that leveraging prosody in your day-to-day speech makes you better understood because the listener doesn't have to devote as much conscious attention to figuring out what you're trying to say.
I remember encountering several of these at the start of Van Vogt's sci-fi classic World of Null-A. I couldn't figure out if it was deliberate in any way, or just an annoying accident.
I don't remember particular sentences like that, but it wouldn't surprise me if it were at least partially intentional. Ambiguity is one of the main components of hypnotic inductions, and van Vogt was a lay expert on hypnosis. (He co-authored "The Hypnotism Handbook" with Charles Cooke, for example.)
It's said that van Vogt would sometimes rely on unconscious processing and "dream logic" (for want of a better term) in his writing processes, so those sentences you noticed might well have been both "accidental" and unconsciously intentional.
>The cotton clothing is made of grows in Mississippi.
Yet the most general pattern seems to be a simple matter of loading too many words into the subject clause [1], e.g.:
>The car driven past the barn crashed.
We don't reach the verb until the last word, which not only opens the door to ambiguity but also produces flaccid prose. You will produce tighter, sharper sentences by front-loading the predicate verb.
[1] - edited from "too many clauses" as per telemachos's astute reply.