1. #1

    Implied Quantifiers: Group X is Z - All or some?

    Kind of random topic here but I was wondering what everyone thinks when they hear someone makes a claim about a group but doesn't quantify it beforehand. On a grammar/logic/language level (rather than the actual content of what is being said), is it the case that someone means All [group] has attribute Z? To pick a non-political example: I could say "Detroit Red Wings fans are great people". This could mean that:

    1) all red wings fans are great
    2) some/most red wings fans are great
    3) what distinguishes red wings fans from other fans is that they're great.

    Hopefully this isn't too pointless, lol. No, this isn't for homework, I just find a lot of these claims going around in politics and the dialogue might be made more clear if we had some understanding of what we mean when we leave the quantifiers up for interpretation/assumption.

  2. #2
    From a purely logical standpoint, it definitely sounds like 1)

    But in terms of actual conversation, I think most people would understand it as most or typical. So more of a mix between 1) and 2).

    But seriously, it only takes an extra word to put a quantifier on something, so I wish people would do that to reduce confusion.
    Quote Originally Posted by Zantos View Post
    There are no 2 species that are 100% identical.
    Quote Originally Posted by Redditor
    can you leftist twits just fucking admit that quantum mechanics has fuck all to do with thermodynamics, that shit is just a pose?

  3. #3
    Quote Originally Posted by Garnier Fructis View Post
    But seriously, it only takes an extra word to put a quantifier on something, so I wish people would do that to reduce confusion.
    Yes. I was guessing that when you define a word then the definition follows through necessarily. Like "Swans are white" - means all swans are white; that's what makes it a swan. I guess there's more nuance in a very philosophical sense because you can't know by definition that swans will always be white. One could turn out black and then your phrase would have to mean "most swans are generally white" or "swans are white, with some exceptions of course".

    - - - Updated - - -

    Quote Originally Posted by Connal View Post
    I think this is just primed language at work. Trump is the master of that. And most people fall for it. To others it is greating. Another person that does this a lot is Gordon Ramsay. "Try the most wonderful beef wellington!"
    Hyperbole is quite annoying. And Trump is a good example. He can make the claim "Mexicans are bringing in drugs and violence" - but then since he has the quantifier implied rather than explicit (to be fair he says some are good people) he can dance around the all/some/many/most he's hidden (although a bit more obvious what he really means since he's said some are good people). It's powerful as a demagogue to put out an unquantified meme that is supported by confirmation bias.
    Last edited by Kraenen; 2017-02-13 at 06:09 PM.

  4. #4
    The second one, you cannot clearly refer to every single individual.

  5. #5
    Herald of the Titans
    10+ Year Old Account
    Join Date
    Mar 2012
    Posts
    2,545
    Logically, I agree from a sentence breakdown standpoint it definitely means #1.

    But in most political usage today, it's often used to generalize all members of a group as having a certain quality even though almost always that generalization is incorrect.

  6. #6
    Quote Originally Posted by Connal View Post
    I honestly wonder though, if doing this actually works for some people. It is akin to the placebo effect. If you think you are going to be eating the most wonderful beef wellington, your mind tends to make you think that it is exactly that.
    Well suggestion + authority + ethos = a pretty powerful and convincing combination. A lot of people, when they've heard that Ramsey is just this superb, authoritative chef, would probably rate his beef wellington higher than when doing a blind taste test. This is probably an easier example since we can run a simple experiment.

  7. #7
    It's all about context. In some situations it might mean "all"..in others it might only mean "some" or "most".
    “The biggest communication problem is we do not listen to understand. We listen to reply,” Stephen Covey.

  8. #8
    Only because of the way I think.

    To me when someone says this I break it down internally. Basically, Red Wings fans, in the niche of being a fan to a team are great fans. Most fans of the Red Wings are loyal to their team and generally support that team through good and bad. This then breaks down inside again, to this can't mean all of the them, but either all this person has met or a decent majority, usually backed up by dedication over the years, sellout crowds in good times and bad etc ( I normally research or think about this from personal knowledge of my own). Then determine if the statement seems factual or not, if it does I normally am willing to fact check it against facts or this persons normal credibility. Strangers rarely get a benefit of the doubt. If it is information I have zero knowledge on I give a slight leaning too true unless the statement sounds baseless or totally false on its face. Then I normally again fact check it, but I'm the person who looks everything up, even jokes I don't get or references, then I normally view said reference material or read the primary source etc.

    Also, compare it to sociological and societal norms I know. ie all people can't be one thing, people are an array of ideas, concepts, niches, likes/dislikes etc. Most blanket statements I realize are perceptions of the person on a group as a whole and not completely true, while possibly maintaining bits of truth.

    Then again, this is very personal to me as I break things down and over internalize all things. It is a personal problem. Do most people do this? No. Most people are sane and have better shit to do than over analyze everything they see and hear. As I said, I have problems. Over analysis and data/knowledge seeking being one of them, and it has no bounds from useful to useless. Its an OCD thing I think, dunno, therapist says OCD/Control issue.

  9. #9
    Quote Originally Posted by Raybourne View Post
    Kind of random topic here but I was wondering what everyone thinks when they hear someone makes a claim about a group but doesn't quantify it beforehand. On a grammar/logic/language level (rather than the actual content of what is being said), is it the case that someone means All [group] has attribute Z? To pick a non-political example: I could say "Detroit Red Wings fans are great people". This could mean that:

    1) all red wings fans are great
    2) some/most red wings fans are great
    3) what distinguishes red wings fans from other fans is that they're great.

    Hopefully this isn't too pointless, lol. No, this isn't for homework, I just find a lot of these claims going around in politics and the dialogue might be made more clear if we had some understanding of what we mean when we leave the quantifiers up for interpretation/assumption.
    Detroit Red Wings fans are great people
    In NST, If we take sum of all people (A), the team (B), 0 and 1 as the domain set, then we can take(*) ( operation which will determine if a member is enthused towards Detroit Red Wings) and ( ×) ( an operation which would result in relative greatness of the members) as operators thusthe premise statement would be:

    ∀a,c∈A & b∈B | a*b=1 ⇒ a x c > 0 ;
    Now for the 3 other statements:
    1) ∀c,d ∈A & b∈B | d∈ {x, x∈A| x.b=1} ⇒ d x c > 0;
    2) ∀a,c ∈A & b∈B | ∃a: a*b=1 V a x c =0;
    3) ∀a,c∈A & b∈B | a x c >0 ⇒ a*b=1 ;

    Now (2) is clearly false considering the premise. But the interesting part is that if you take greatness as a boolean both 1st and third statements can be derived from the premise. Otherwise, premise would be false to begin with.

    So yea, taking the premise as true, automatically makes the whole system biased towards Detroit Red Wings fans, despite the existence or lack there of a quantifier.

    As you see evaluating a statement is easy, hard part is to come to an agreement on logical structure.

    This post is of course full of syntax errors, it's hard to type on a phone. However I hope I could explain my self.
    Last edited by HumbleDuck; 2017-02-13 at 08:49 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •