Ambiguous plurality

[VictorGijsbers]

I have two questions about the “amibuously plural” property in Emily Short’s extension Plurality. The first is: why was it designed this way? The second is: is there an easy way to fix it to do what I want, or do I need to do a wholesale replacement of section 1?

To remind you, this is what ambigous plurality is for:

We also create the term “ambiguously plural” to refer to items such as “pair of socks”, “set of chessmen”, “crew of pirates”, etc., which the player might reasonably refer to as either “it” or “them”. These items will then respond equally well to “it” or “them” in input.

However, Plurality does more: it also defines a phrase “acts plural”, which is true for an object if it is either plural-named or ambiguously plural. The new grammar tokens introduced by the extension ([has-have], [is-are], the extended [s]) then choose the plural form if the object they apply to acts plural. This means that ambiguously plural things get the plural verbs.

First question: why? All the examples given (“pair of socks”, “set of chessmen”) should get singular verbs, not plural verbs. This is true for the objects in my game as well (“swarm of daggers”, “pair of nunchucks”). In fact, I can’t immediately think of anything that shold be ambiguously plural and have plural verbs applied to it. Am I missing something here?

Second question: I could fix this by changing the “To decide whether (item - an object) acts plural:” code in Plurality, but that involves an ugly substitution of the entire section 1. If anyone sees a better solution, please let me know!

[Dannii]

I think you’re right, the examples I can think of should take singular verbs too.

If you define a phrase with the exact same wording as one in an extension, it is supposed to overwrite it. It’s not reliable when you do that in another extension due to ordering issues, but if you do it in the main source file it should be okay.

It’s probably worth filing this as a bug. (Though perhaps wait a while to see if Emily sees this first.)

[dahaiou]

Ahem, if it helps, here are a few example sentences found by googling for “a pair of socks”, “a set of chessmen” and “a crew of pirates” respectively:

Her local paper reports she was ‘dumbfounded’ when a pair of socks were discovered in the pram.

You really should consider hanging out with Aubrey," Chloe commented; laughing and ducking as a pair of socks were thrown at her.

… for how could a tramp know that a set of chessmen were worth carrying off?

A set of chessmen were presented me about three years ago, which go into the compass of one-third of an inch square

During the first presidency of Nicholson, a crew of pirates were captured on the coast of Virginia;

He hid behind the black flag and peered around. A crew of pirates were spreading out on the ship and Terence curled his nose up in disgust …

For each of these you can also find examples of verbs used in the singular, though for “a pair of socks” the plural uses seem to be in overwhelming majority.

[VictorGijsbers]

Surely those sentences are just wrong? I mean, “a pair” is singular, right?

[vyznev]

@VictorGijsbers

Surely those sentences are just wrong? I mean, “a pair” is singular, right?

I have a vague suspicion that this may be, at least to some extent, a difference in usage between British and American English. In fact, Wikipedia seems to agree:

@Wikipedia

n British English, it is generally accepted that collective nouns can take either singular or plural verb forms depending on the context and the metonymic shift that it implies. For example, “the team is in the dressing room” (formal agreement) refers to the team as an ensemble, whilst “the team are fighting among themselves” (notional agreement) refers to the team as individuals. This is also British English practice with names of countries and cities in sports contexts; for example, “Germany have won the competition.”, “Madrid have lost three consecutive matches.”, etc. In American English, collective nouns almost invariably take singular verb forms (formal agreement). In cases where a metonymic shift would be otherwise revealed nearby, the whole sentence may be recast to avoid the metonymy. (For example, “The team are fighting among themselves” may become “the team members are fighting among themselves” or simply “The team is fighting.”)

Also, it’s possible that the examples mentioned in the documentation may have been referring to something like the following:

The Testing Chamber is a room.
Some socks are in the chamber. The indefinite article is "a pair of".
Some chessmen are in the chamber. The indefinite article is "a set of".
Some pirates are people in the chamber. The indefinite article is "a crew of".

which will cause the socks to be referred to as “a pair of socks” in things like room descriptions, but simply as “the socks” in contexts where a definite article is called for.

[Dannii]

I think it depends on whether or not it is definite or not. “A crew of pirates were captured” is fine, “the crew of pirates were captured” is questionable, “was” would be much more natural.

[Peter_Piers1 (Peter Piers)]

My two cents: grammar is fine and we should all write correctly.

However, if it feels more natural for the player, at a certain instance, to consider a singular item as plural, because in their heads it makes more sense (“I’m interacting with the bees → them. I’m interacting with the swarm → it”), then not allowing the player to do so in the name of “it’s incorrect” is a silly barrier.

It’s like forcing a british player to write “armor”, or an american to write “armour”.

I’m reminded of something I read somewhere (one of Jimmy Maher’s articles?) where a person from Magnetic Scrolls was asked why a certain sytax wouldn’t work in their parser where it worked on Infocom. The person grew very suspicious and asked, “Why should it work? It’s not grammatically correct”. I think the issue was with a preposition. I also seem to remember the issue was more with american/british english, and in the end they grudgingly started accepting that syntax. Power to them.

EDIT: Ah, here we go.

English is not English. That became obviously clear at a press conference in London, which was attended by Dave Lebling (Spellbreaker, Starcross) from Infocom and Anita Sinclair (The Pawn) from Magnetic Scrolls.

While chatting we also discussed problems Infocom fans have with the parser of The Pawn. German Infocom User group pointed out that The Pawn does not understand “look into”.

Asking Anita about that, she became suspicious for a very short moment, only to counter: “Why should our parser understand this? Look into is completely wrong in English grammar”. However, Dave Lebling answered: “In American English “Look into” is a common term.” Anita agreed to support “Look into” in future games, even if it conflicts with grammar rules. Vice versa Dave Lebling promised to show more consideration to English people in his next adventure, because Anita immediately complained about a scene in Spellbreaker, in which the parser understands American English, but not British English.

– Boris Schneider (Happy Computing, June 1987)

[VictorGijsbers]

Peter, this discussion is about the prose generated by the game, not about accepted player input. There, of course, one can simply be inclusive.

[Peter_Piers1 (Peter Piers)]

Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaah.

Nevermind, then.

[PaulS]

In actual use (I can only speak for British English) these ambiguous plurals really are ambiguous. In British English usage “nouns of multitude” can be treated as singular or plural, as the writer prefer, though there are common usages. Practices vary even in tiny and pedantic communities: when I first became a barrister it was consider wrong to refer to a company as singular in one division of the High Court, and wrong to refer to it as plural in another!

Here’s Fowler, a great authority, from 1908

Though nouns of multitude may be freely used with either a singular or a plural verb, or be referred to by pronouns of singular or plural meaning, they should not have both (except for special reasons and upon deliberation) in the same sentence; and words that will rank in one context as nouns of multitude may be very awkward if so used in another.

For what it’s worth as a native speaker (and therefore, of course, illogical and often wrong – but at least natural) it would never cross my mind to refer to a pair of socks in the singular. “What colour is this pair of socks? It is red.” That sounds unnatural – the speech of a foreigner who has learned from a book, not a native. The same with, say “dozen”: “I want a dozen eggs please.” “What colour would you like it to be?” No. No. No.

On the other hand with words like “committee” and “government”, although I might in speech use the plural, I’m pretty sure I would use the singular in writing … mostly, though it might depend on whether I was emphasising unity or diversity.

In the original case, “a pair of nunchucks” should most certainly be plural to be idiomatic in British English; not that the singular is wrong – it’s just not natural. But the swarm of daggers, when it acts as one, should probably be singular. I daresay that makes the coding yet more difficult.

[Felix_Larsson]

Perhaps you could introduce the property of being “ambiguously singular” and add rules to ‘notice the plurality of’ ambiguously singular thing, like Plurality does for ambiguously plural things (in Section 4 of the extension). Then the reader should be able to refer to ambiguously singular things as either “it” or “them”, even though they don’t ‘act plural’.

[matt_weiner (matt w)]

I’ve just written the bit of code that I think introduces the “ambiguously singular” property, I think. It seems as though the code that allows ambiguously plural things to be referred to by both pronouns is all in section 4. The reason that they use plural verbs etc. is because they’re defined as “acting plural” by a phrase in section 1 that checks whether an object is plural-named or ambiguously plural. So I defined the third value “ambiguously singular” and redirected the checks for ambiguous plurality in section 4 to ambiguity tout court. Ambiguously singular things automatically don’t act plural (since that’s not part of the check), so they should get all the singular object behavior that plurality usually gives you.

Here’s the code! Let me know if it seems buggy (really, do, because I’m dropping it in my project without that much testing).

[code]Section - Ambiguously Singular (in place of Section 4 - Enumeration in Plurality by Emily Short)

[This adds a new option, “ambiguously singular.” Ambiguously singular things act like ambiguously plural things for the purposes of all the pronoun-setting stuff that is going on in this section, but they don’t “act plural” for the purpose of the things that are happening in section 1 of Plurality, so they should use singular verbs etc.]

A thing can be ambiguously plural, ambiguously singular, or ordinarily enumerated. A thing is usually ordinarily enumerated.
Definition: A thing is ambiguous if it is ambiguously plural or it is ambiguously singular.

This is the update ambiguous pronouns for actions rule:
if the noun is a thing and the noun is ambiguous:
notice the plurality of the noun;
if the second noun is a thing and the second noun is ambiguous:
notice the plurality of the second noun.

The update ambiguous pronouns for actions rule is listed after the set pronouns from items from multiple object lists rule in the action-processing rules.

After printing the name of an ambiguous thing (called the suspect) (this is the notice plurality of printed ambiguous object rule):
if the manual pronouns option is active:
do nothing;
otherwise:
notice the plurality of the suspect.

To notice the plurality of (suspect - an object):
(- PlugPlural({suspect}); -)

Include (-

[ PlugPlural obj;
SetPronoun(‘them’, obj);
SetPronoun(‘it’, obj);
];

-)

Use Manual Pronouns translates as (- Constant MANUAL_PRONOUNS; -)

To have the parser notice (o - an object):
if o is ambiguous, notice the plurality of o;
otherwise set pronouns from o.

[The following is a legacy item in case people are still using it in their code, but “set pronouns from…” will work just as well.]

To notice (o - an object) boringly:
(- PronounNotice({o}); -)
[/code]