"Not all___ are like that!"

How do you counter this argument? Is it valid, or a way to derail the conversation?

Other urls found in this thread:

youtube.com/watch?v=Ry3NzkAOo3s
twitter.com/NSFWRedditImage

ol' saying nd a smug face, if they're not going to take this conversation seriously why would i?

Not all Nazis gassed Jews

you cant dumbass

This is one of the more common ways to counter an arguement. Take advantage of the variables infinite possibilities and you can use their logic against them.

If they are a true liberal then they would be offended by things like
Not all Religions are false, unless you provide proof otherwise then you are marginalizing a group you know little about.

This, and if they try to say that the silent ones approved of gassing jews, then you move on to say the muslims who are silent are just as guilty as the ones who commit terror. To argue with a leftist you must attack using their feelings against them. Make them question their own ideology. Otherwise it's pointless to argue with a leftist.

Easy. The peaceful majority of a group is irrelevant. Most Nazi Germans were probably peaceful, as were Chinese under Mao, soviets too, etc etc

All x may not be like that, but think of it like a graph, muslims on average are more violent even though there are non violent muslims

what a shitty argument

...

s k i t t l e s

Then why are horses, cats and dogs the same everywhere they are found on this earth?

>not all terrorists are Arab but they are

Maybe clarify also that that doesn't mean 100% but rather a very large portion.

youtube.com/watch?v=Ry3NzkAOo3s

"Not all___ are like that!" = ">Sup Forums is one person"

We're setting the rule, not the exception

it's a way to derail the conversation.

regardless of whether or not all X do Y, you're complaining about the X that do indeed do Y and are prominent

e.g. feminists that are essentially anti-men, anti-white, anti-straight

just because all of x is not like that, doesn't invalidate the fact that most of y is x

"So do you think all whites should apologize for slavery 200 years ago when it was banned in half the country and only 2% of Southerners out slaves, which included other blacks and Jews?"

If they say yes or start rabbling on about there being "many factors" or some shit then just laugh at them, call them braindead retards, and walk off because arguing with them would be a complete waste of time

1% of a group not doing something doesn't negate 51% of a group that does.

If most women do something, it's safe to say "women do..." That's not to imply that 100% of "women do...", but if most women do it, if a majority do, if 51% do it, then odds are, any given women does it. The likelihood is greater. The probability is greater.

51% is closer to 100% than 0%.

Saying NAWALT (not all women are like that) is a result of black-or-white thinking, or 0-or-100 thinking. They think if a number is non-zero, it's meaningful (but 1% of people don't do that!) and they think if a number is non-100, it's meaningful (but 100% of people don't do that!). They have no concept of probabilities, odds, numbers between 0 and 100.

>Not all snakes are venomous
>let me put you in a tank with various snakes

>Not all muslims are terrorists
It barely even took a dozen to bring down the world trade center, killing 2000+ people. If you're going to tell me we shouldn't be worried about a 1:100 kill ratio, you shouldn't be allowed to contribute to the conversation.

and then the discussion ends with a "c'mon user, be serious" or "it's not the same and you know it" with them learning nothing and thinking you're a loon

How about how certain dog breeds are more predisposed to killing than others? We can't talk about that!

Using this forever, thanks.

>Then why are horses, cats and dogs the same everywhere they are found on this earth?
They're not

yep this came to mind

This. Exactly.

Speaking of generalizations, if one snake bites you, it's better to err on the side of caution and be wary of all snakes, instead of taking snakes on a case by case basis, trying to individually judge snakes.

Stereotypes evolved because it makes sense evolutionarily to assume the worst and be wrong (run from a noise, it was only a rock falling) than to assume the best and be wrong (that watering hole looks safe, oh shit i'm being eaten by an alligator).

It's better to assume a snake is venomous and could kill you (and be wrong), than to assume a venomous snake is safe and won't kill you (and be wrong).

It's better to assume a mushroom is poisonous and could kill you (and be wrong), than to assume a poisonous mushroom is safe and won't kill you (and be wrong).

Ask them to explain the difference.

"enough of them are like that to have an effect"

How dare you cast doubt on the equality of horses! Small ponies are severely underrepresented as winners of the Kentucky Derby! We need affirmative action for horses now!

Youre making a generalization OP. So the argument doesnt actually counter your argument.

OP could reply, the existence of a minority does not diminish the existence of a majority.

When I hear that argument, I usually say that if 1 person wants to murder you and 9 people are watching this and do nothing to stop the murderer, then they are also guilty for secretly wanting you dead but not having the balls to do it themselves.

It doesn't take all 10 of them for you to get killed. It only takes 1 person willing to do it, and 9 person not to stop him to get you killed.

Being a human being means you have morals, and watching someone performing a vile act should make you feel sick inside.

It only takes 10 extremists to convince 1000 normies to generally support their views, even if they don't completely agree with them.

It only takes 1% of men to protect the country and be soldiers. That doesn't mean that everyone has to be a soldier, but we do encourage people to become soldiers, because while we know that not everyone will be, shooting for the moon lands you on the stars and the 1% gets filled, thus the country will be protected adequately.

Is she still alive or did she die of an heart attack already?

My reply isn't bait, but the woman answering did not answer the girl's question in any way. The Arab girl asked 'Why are we fighting jihadist ideology using weapons? Isn't there a better way to do so, e.g. try to fight jihadism using an idealogical approach rather than a violent one?'

The answerer never spoke to that at all; she instead gave an answer on why you shouldn't use the argument 'Not all Muslims are bad.' Granted, the Arab girl seemed to imply that before asking her question, but that wasn't her question at all.

Just goes to show you that so many people choose to ignore what people say and launch into what they think they say. Right wing retards on pol think she just slayed that Arab girl, when really the person never even answered the question.

The exception does not disprove the rule

Bring up bleach.
>According to the FDA, anything more than a tablespoon of bleach in a gallon of water makes you sick. That's 200/1000000. Point-zero-two percent. Or, let me put it this way... 1 non-moderate per 5,000 migrants.

That's another big one, use the term "non-moderate," instead of radical.

"Third world ideologies and problems require third world solutions."

In the movie Thank You for Smoking, Aaron Eckhart plays a slimy spokesman for Big Tobacco. When he gets asked questions, he likes to avoid them, and then talk about something else, saying "here's what I want to talk about." Just classic spin.

see

the gas chambers weren't real you fucking kike shill

It's % based. Example: Muslims account for 80% of all terrorist attacks ( or whatever the number is). The "not all" argument is a huge fallacy.

It's valid but shallow, both sides should be looking to proper statistical analysis before they draw conclusions.

What proportion of X demographic acts in Y way? What is the actual impact of Y-type actions on Z demographic, as opposed to Z's perceived impact?

Only once a reliable statistical breakdown has been thoroughly examined can either side start proposing what they believe to be appropriate measures to counter Y-type behaviour in X demographic. Of course, in the current heated political climate it's unlikely that either X or Z are going to be willing to compromise, but until reliable numbers are available the whole conversation is even more pointless.

Depends on point. Usually, 'thats a non problem, we don't talk about cancer in terms of being a minority's or you could say it has to do with 'the environment is hosting the means to the end'. You can also break it down further from there and be more specific

I just rewatched the video twice and she never says this, so if You have a timestamp please link it. That would be a proper response in my opinion

Exception makes the rule.

Pick a cat off the street, put them in a filled bathtub with just your hands. Who expects a random stray cat to like water?

Yet SOME *do.

Oh, I wasn't saying it was from the video, sorry.

There is no counter because the argument is valid.
If you say "ALL X are Y" then all you need to disprove the statement is provide a single counter example: "There exists X that isn't Y". Logically, that's it.
The moral: don't make sweeping generalizations because they are hard to justify logically. Reason better.

#Notall is a non sequitur. It has jack to do with the argument.

> Not all Muslims are terrorists or extremely awful religious fundamentalists
No, but almost all terrorists and extremely awful religious fundamentalists of Muslims.

> You can't judge group X for action Y, think of individuals
Thinking collectively can be good for a society. Not all muslims may be bad in Germany or France, but if there were no Muslims in those countries then they wouldn't be experiencing terrorist attacks.

>52% total

Not all Jews died in the holocaust

it's tough

splitting hairs is a classic liberal actic, they hope to turn a straightforward debate into a bog of definition-mongering, exception making, and they will try to force you to detail the smallest nut and bolt

counter with "exemptions can be made as they present themselves, but something must be done about the issue"

or "the issue is too big to seek a perfect solution, delay is hurting us constantly and while you are fiddling with the finer points rome is burning"

People rarely says "all x are y." They use a plural word, like women, men, dogs, cats, etc.

The point of contention comes from the interpretation of a plural.

If someone says "women are golddiggers", women means more than one woman, so at least 2 women. By saying "women are golddiggers" does that mean ALL women are golddiggers? Does it mean 100% of women are golddiggers?

If 51% of women are golddiggers, if most women are golddiggers, isn't it safe to say that "women are golddiggers"? That women tend to be golddiggers? That the likelihood of any woman being a golddigger is greater than the likelihood of her not being a golddigger? That the odds are that any random woman you meet is a golddigger?

If 51% of a group do something, a generalization holds. The odds are closer to 100% than 0%.

But that makes my job as a bigot harder! I became a bigot because I didn't want to think more.

Not all trucks deliberate rams into people.

this is good

>How do you counter this argument?
You can't because it's not an argument.

thats true

could you even call it a holocaust in that case

It depends entirely upon the argument you made. If your argument hinges upon the idea that a group of people acts entirely alike, then that argument is completely valid.

>Claim
Not all blacks are criminals
>Counterclaim, retard verson (pictured in OP)
hurr muh NAXALT, I've never heard the phrase 'black swan' before
>Counterclaim, better version (not pictured)
The existence of a black swan does not preclude the existence of white swans, it merely implies that we need to broaden our definition of 'swan.' Our program of racial classification is in no way made more difficult by the existence of black people who do not violate laws. An albino nigger, for instance, is still a nigger; he is simply an abnormal specimen and wholly worthy of note as an example of the beauty of biodiversity--not that I'd want to share a continent with him.

Those digits don't lie

That's a cute truism, but it's terrible policy.

Western and Russian invasions of the Middle East and physical wars on Islam have done nothing but destroy anything approaching a valid and stable first-world government in the target countries, which has generated a growing power vacuum that is swiftly filled by religious extremists and warlords, making the problem worse.

When these countries weren't warzones, they didn't export terrorism at all, and religion and state were often (although by no means always) separated in a way that was in line with what many today would see as a purely Western standard.

The military-industrial complex loves the current status quo, of course, because it equates to increased sales of weapons to both sides in the affected area, as well as security technology in the invading nations, not to mention the countries that are forced to deal with waves of refugees from the war. So there will always be a powerful lobby in Western governments that argues for continued military interference in countries where zealotry and lawlessness are on the rise, but that lobby is not interested in reducing either factor.

Generalizations work better if backed by statistics. Most Muslims support sharia law. Most blacks come from single parent homes.

Arab countries were doing fine and weren't exporting people when secular dictators were in charge of their countries. Nasser in Egypt, Sada'am in Iraq, the Shah in Iran and even now Iran doesn't export terrorists or do much in the way of citizen terrorism, it's all their government, Ba'athists in many places like Syria, etc. there's a long list of Secular Arab Nationalists who didn't fuck things up and got along fine with Western leaders and norms and we and Israel have done a lot to fuck things up.

>not all niggers are black.

>not all Jews are evil scum.

>not all Australians are shitposters

51% is a majority. A majority can also be bigger than 51%. Notice how I didn't say "the 51%", I said "51% of a group", when you can tell a majority exists. There's a difference between "a 51%" and "the 51%." That accounts for unknowns. "The" suggests definitive, solid, immutable amounts, total knowledge.

Plus, knowing the actual figures, the actual percentages is often unknowable. However, determing MOST is much easier. You figure out what happens most of the time. You figure out what basket a person falls in most of the time.

But the point is that the existence of a minority does negate the existence of a majority.

You have to realize at that point in the conversation,the person who said that can not be reasoned with, and fully believe all races are created equal, move on,

>not all women are hypogamous gold digging cunts.

not all shit is brown

>not all Australians are cunts.

Do you treat all black widow spiders on a case by case basis?

It's better for survival for a lifeform to assume the worst and be wrong (that thing is a threat, death doesn't happen), than to assume the best and be wrong (that thing isn't a threat, dead).

>not all Canadians are stupid.

Ask them why you should care about the supposed "peaceful majority"

>but most blacks/ muslims are PEACEFUL!!!

Ask them how being only mindful of the "peaceful" majority is serving Europe and American whites right now. It only takes a few of the violent minority to fuck my shit up and ye peaceful majority simply say "but I'm not them!" And do nothing else.

So fuck them.

It's pretty disingenuous to have accurate enough statistics to be able to see that exactly 51% of a demographic behaves in a specific manner, and then remove that number from your argument and instead switch it out for the word "most".

If anything, the word "half" would be much more appropriate and less likely to result in misunderstandings, since 51% is a damn sight closer to 50% than it is to 100%. The same obviously applies to using "half" instead of "few" when the figure is 49%.

Obviously it's a waste of time to try and decide on specific numbers where certain weasel words are appropriate or not, which is why it is always preferable to use the exact number if it's available (and it probably is, in the modern marketing age a lot of work goes into studying demographics and their behaviour).

>Americans are stupid..sorry

If someone says "Muslims" do they mean 100% of Muslims, or "more than one Muslim"?

There's a reason the word "all" exists. It's a qualifier.

If someone says "All Muslims do...", you can respond "Not all Muslims do..."

If someone says "Muslims do...", a response of "Not all Muslims do..." assumes they mean ALL when all wasn't said.

>not all muslims are self hating brainwashed suicidal morons.

Thanks Ben Shapiro

>How do you counter this argument? Is it valid, or a way to derail the conversation?

It's a way to derail the conversation.

Throw back at them your choice of the following.

Hit them with the racism angle at first.

Not all KKK members lynch blacks.
Not all nazi's kill jews.

Then hit them with this zinger.

Not all Russian Roulette players blow their brains out. Care to play a game?

When you say "women are gold diggers", do you mean "some women are gold diggers" or do you mean "all women are gold diggers"? The issue is twofold: you probably mean the former but then argue as if it were the latter.
The statement is ambiguous. Specify.

>If 51% of a group do something, a generalization holds. The odds are closer to 100% than 0%.
American education folks. This is what passes for logical reasoning in America.

Hahaha!

youtube.com/watch?v=Ry3NzkAOo3s

one of my favorite videos on youtube

I never intended for 1% and 51% to be added together.

But 1% is non-zero, more than zero, it establishes existence. I could add in decimal points, but it's a simple analogy.

And 51% is more than half. 51% is most. I could add in decimal points, but it's a simple analogy.

The point is most of the time vs at least once.

51% is closer to half than 100%, but 51% (most) is closer to being true all of the time (100%) than true none of the time (0%). Odds are in the favor of the 51% group. For people betting, it makes more sense to bet on the thing that happens 51% of the time (ignoring risk and reward and payouts). It's better to tune your expectations to the 51%.

I'm saying that when saying "cats have four legs", that doesn't necessarily mean that 100% of cats have four legs (otherwise we would say "all cats have four legs"), but typically, most of the time, odds are, over half of the time, cats have four legs.

Depends on the context.

For example when talking about islam since islaam is an idea

>muh not all muslim
>ok if you are muslim you must accept the Quran, if they don't then they are hypocrites
>what about Christians
>not all Christians
>retort same argument
>the difference here is that Christians don't have a passage permitting them to lie to you about their faith and idea's. Further more Christian theology sees our weakness as normal which is why there are variations of the same passages and we can live relatively peacefully.

If a statement holds true more than half of the time, it's safe to make.

By saying "women are gold diggers" that doesn't mean "all women are gold diggers", otherwise one would say "all women are gold diggers." The word "all" exists and people use it. However, if someone says "women are gold diggers", you can assume they mean "most women are gold diggers", meaning more than half, typically, usually, odds are, generally speaking. If someone responds "not all women are gold diggers", they are countering a claim that wasn't made, they are assuming a plural means 100%.

By saying "cats have four legs" that doesn't mean "all cats have four legs", otherwise one would say "all cats have four legs." The word "all" exists and people use it. However, if someone says "cats have four legs", you can assume they mean "most cats have four legs", meaning more than half, typically, usually, odds are, generally speaking. If someone responds "not all cats have four legs", they are countering a claim that wasn't made, they are assuming a plural means 100%.

And yes, the number 51 is closer to the number 100 (only 49 away) than it is to the 0 (51 away). And something that happens 51% of the time exists, whereas something that happens 0% of the time does not exist.

If something is more likely than not, if something is more probable than not, if safe to assume.

Stereotypes evolved because it makes sense evolutionarily to assume the worst and be wrong (run from a noise, it was only a rock falling) than to assume the best and be wrong (that watering hole looks safe, oh shit i'm being eaten by an alligator).

It's better to assume a snake is venomous and could kill you (and be wrong), than to assume a venomous snake is safe and won't kill you (and be wrong). The startle reflex did not evolve to judge threatening lifeforms on a "tolerant" case by case basis, because that can get you killed.

According to FBI crime statistics, white people accounted for 69.3% of crimes in the USA. Therefore, white people, are, in general, criminals! It is safe to assume ALL whites are criminals!!! White people are, in general, criminals!

In 2010, females accounted for approximately 50.8% of the US population. Therefore, Americans are, in general, female! Which means all Americans, in general, use tampons and have periods!

Boy, reasoning is easy when you're stupid.

See

Like I said: reasoning is easy when you're stupid. I don't expect much from Sup Forums, but this is surprising given your sticky is a big poster of logical fallacies.

'dude not all nazis were bad'

This. Do they know how easy it would be for 20 men to wreck havoc in a busy city?

It's as simple as setting up a kill box outside a busy building and pulling a fire alarm. If even 1 in 1000 is radicalized, its a big fucking problem.

If most crimes are committed by white people in the US, then yes, it's safe to assume if a crime is being committed that a white person did it. Note that that argument depends on the truthfulness of the premise.

If most people in the US are female, then yes, it's safe to assume that if you pick a random person in the US, over time, odds are, they will be female.

I never said most means all.

This whole "NAWALT" bullshit is people thinking "cats do" = "all cats do" rather than "most cats do."

It's like you entirely skipped the part about using "all" and how saying "most" does not mean "all."

Fucking leaf posts.

Nice strawman, no one is saying "all muslims are terrorists".

Besides, why do you need to prove "every muslim is a terrorist" before implementing a reasonable immigration policy.

Hey what's it like being a rocket scientist?

Ben carson you fucking racissssssst!

>white people accounted for 69.3% of crimes in the USA. Therefore, white people, are, in general, criminals!

You are fucking retarded.
By that statistic you can say "ciminals in america are (mostly) white", not the other way around.

It doesn't matter if not all are like that there's still a sizeable amount that are and if you're going to associate with that you need to condemn their actions and keep them under control.

>leaf
Can you check the FBI's way of defining "white"? I'm willing to bet a lot of those white criminals are spics.

Do you know what odds are, you stupid fucking leaf? Ever heard of the terms "usually", "typically", "more often than not", "most of the time", "generally speaking"?

By saying "cats have four legs" that doesn't mean "all cats have four legs", otherwise one would say "all cats have four legs." The word "all" exists and people use it. However, if someone says "cats have four legs", you can assume they mean "most cats have four legs", meaning more than half, typically, usually, odds are, generally speaking. If someone says "cats have four legs" and someone responds "not all cats have four legs", they are countering a claim that wasn't made, they are assuming a plural means 100%. They are assuming "cats" always means "all cats", making the word "all" completely superfluous and unnecessary.

If it helps your puny brain, instead of interpreting the phrase "cats have four legs" to mean "all cats have four legs" or "100% of cats have four legs", leading you to the kneejerk response of pointing to a cat that does not have for legs, you can safely assume they mean MOST cats instead of ALL cats.

Do you know what "in general" means? A generalization is a statement about a group of people or things based on only a few people or things in that group. A proper generalizations holds true over half of the time.

Generalizations also evolved for a reason. It's safer to err on the side of caution and be wrong, than to have a false sense of security and be wrong.