US Officially Blames Russia

[phiwum]

Not quite, it's pretty much the definition of "logical fallacy".

No, this is just flat wrong.

Remember that article on fallacies? Right at the top of the article was a note that they omitted "formal fallacies", i.e. fallacies in deductive logic.

Appeal to authority and other common fallacies are considered informal fallacies, tied almost exclusively to inductive reasoning. (Nonetheless, they are also called "logical fallacies", as Copi does.)

That's what we have statistics or probabilistic reasoning for. But let's remember that your argument for believing the US government/intelligence isn't based on statistics.

Unless one adopts strict Bayesianism, with the arbitrariness of subjective priors, one cannot calculate the probability that swans are white, given a number of sightings of white swans (with no sightings of non-white).

And, of course, it is foolish to pretend this is how we reason. We see a pattern in nature often enough, we conclude that it probably holds generally and see whether we could find a counterexample.

Again, if that were the case then you have to provide us with estimates of likelihoods. Specifically, let E be "the US government & intelligence services say X is true" and H be "X is true", then you should provide us with P(E | H) and P(E | ~H).

I'll do that just as soon as you provide me with

P(All swans are white | I've seen 275 swans, all white).

True but irrelevant for the reason stated earlier. Even if someone accepts certain claims based on certain authorities doesn't mean someone should accept all claims based on all purported authorities.

Ah, ah! Don't forget what you said: appeal to authority is always fallacious and so should never be done. So, I guess you have no opinion about whether Hannibal existed.

Snip rest, since you're changing the topic. Let's settle our introduction to logical fallacies first.

 

[caveman1917]

Again, we are not speaking of deductive proof here.

I am.

It goes on another two-thirds of a page from there, but I don't care to type any more. What is clear is that appeal to authority is not always fallacious, contrary to your claims.

Define "fallacy".

If you'd like to amend that claim

I'm not amending my claim.

and say that the issue is instead whether intelligence agencies should be treated as reliable authorities, that would be a different matter, not one easy to settle and which comes down to a difference of opinion.

Not "instead" but "in addition to". If you want to use a probabilistic argument rather than a deductive one, which you seem to, then that would be the first issue, yes.

That's nonsense, a gross misunderstanding of informal logic/critical thinking.

There's no such thing as informal logic.

 

[TragicMonkey]

I can't prove a negative

Ah, but can you not not prove a negative?

A double negative! I'm afraid you gave yourself away. -- Wadsworth

 

[caveman1917]

As I said, it is a kind of logical fallacy. It does not represent them all.

You are plain wrong.

 

[phiwum]

I am.

Then you're very confused, because appeal to authority is not a fallacy in deductive logic, since it occurs only in the context of inductive arguments.

Define "fallacy".

Copi's definition will do: a type of incorrect argument.

I'm not amending my claim.

Oh? What about Copi's text? Did I type that for nothing? Why not look up WP's article on Appeal to Authority and see that it is commonly recognized that not every appeal to authority is fallacious?

Or do you just know better than logicians like Copi or those cited in WP?

Not "instead" but "in addition to". If you want to use a probabilistic argument rather than a deductive one, which you seem to, then that would be the first issue, yes.



There's no such thing as informal logic.

Strange, I've taught informal logic many times as part of a critical thinking class. But go on, keep making such silly pronouncements.

 

[caveman1917]

No, this is just flat wrong.

Remember that article on fallacies? Right at the top of the article was a note that they omitted "formal fallacies", i.e. fallacies in deductive logic.

Appeal to authority and other common fallacies are considered informal fallacies, tied almost exclusively to inductive reasoning. (Nonetheless, they are also called "logical fallacies", as Copi does.)

A "formal fallacy" is a fallacy which makes an argument formally invalid (ie based on its logical form). For example (P => Q) => (~P => ~Q). Note how it doesn't matter what P or Q here are, the problem is solely with the form of the argument.

A "logical fallacy" is an argument such that the conclusion doesn't follow from the premises, so I suppose "formal fallacies" are a subset of "logical fallacies". The main point in all this is, however, that an appeal to authority is a fallacy in that its conclusion doesn't follow from its premises.

Unless one adopts strict Bayesianism, with the arbitrariness of subjective priors, one cannot calculate the probability that swans are white, given a number of sightings of white swans (with no sightings of non-white).

This is false, likelihoods are independent of priors.

And, of course, it is foolish to pretend this is how we reason. We see a pattern in nature often enough, we conclude that it probably holds generally and see whether we could find a counterexample.

Who cares about how we reason? We constantly reason incorrectly.

I'll do that just as soon as you provide me with

P(All swans are white | I've seen 275 swans, all white).

Rather P(I've seen 275 swans, all white | All swans are white) which is 1.

Ah, ah! Don't forget what you said: appeal to authority is always fallacious and so should never be done. So, I guess you have no opinion about whether Hannibal existed.

I didn't say it should never be done, I said that it is always fallacious. There's a big difference between "fallacious argument" and "false conclusion".

 

[Childlike Empress]

I have no example handy. Prove a negative or at least explain why that sentence is in the report.

I can't prove a negative, but I've been hearing this same strawman on conservative media quite a bit based, largely I suspect, on a misunderstanding of the word "hacking". I imagine it was added because of that.

I suspect you'll be looking for quite a while before you do find an example. Thus, a strawman.

Yeah, you can't prove a negative, but you could have easily dropped a word about my following request, instead you are deflecting. I did a simple google search for russia meddling voting machines leading me to hits

#1: Washington Pest: Secret CIA assessment says Russia was trying to help Trump win White House

[...] Specifically, the White House wanted congressional leaders to sign off on a bipartisan statement urging state and local officials to take federal help in protecting their voting-registration and balloting machines from Russian cyber-intrusions.

Though U.S. intelligence agencies were skeptical that hackers would be able to manipulate the election results in a systematic way, the White House feared that Russia would attempt to do so, sowing doubt about the fundamental mechanisms of democracy and potentially forcing a more dangerous confrontation between Washington and Moscow. [...]

#2: New York Slimes: U.S. Officials Defend Integrity of Vote, Despite Hacking Fears

The Obama administration said on Friday that despite Russian attempts to undermine the presidential election, it has concluded that the results “accurately reflect the will of the American people.”

The statement came as liberal opponents of Donald J. Trump, some citing fears of vote hacking, are seeking recounts in three states — Wisconsin, Michigan and Pennsylvania — where his margin of victory was extremely thin. [...]

Both post election, both insinuating that there are legitimate fears that Russia directly hacked the election, and both containing statements of officials, like I said, those who can be hold accountable for their words, that this is unlikely.

And now we have it in that "report" explicitly, and the reason they state it is that it was (and will be in gullible circles) part of the campaign. Thanks for playing.

EDIT: Strike that last sentence or at least don't take the tone too seriously (comes with the territory), Upchurch. I think you're a swell guy and could be honestly suffering from a bit of PTrumpSD™. I have even friends here in Germany who haven't recovered yet.

 

[phiwum]

A "formal fallacy" is a fallacy which makes an argument formally invalid (ie based on its logical form). For example (P => Q) => (~P => ~Q). Note how it doesn't matter what P or Q here are, the problem is solely with the form of the argument.

A "logical fallacy" is an argument such that the conclusion doesn't follow from the premises, so I suppose "formal fallacies" are a subset of "logical fallacies". The main point in all this is, however, that an appeal to authority is a fallacy in that its conclusion doesn't follow from its premises.

No, this definition reduces it to purely deductive arguments once again. You ignore the fact that much of the time we use inductive arguments, and rightly so. We appeal to analogy, for instance, or we draw conclusions about white swans of the sort already mentioned. Most of our informal fallacies are about bad inductive arguments, arguments whose premises do not suffice to ensure that the conclusion is probable.

I don't know why you insist on missing this basic point. Perhaps you should look up inductive argument.

This is false, likelihoods are independent of priors.

Only in the limit, but not at any particular finite number of observations.

Who cares about how we reason? We constantly reason incorrectly.

I have nothing to say to this.

Rather P(I've seen 275 swans, all white | All swans are white) which is 1.

And do it again with the negation of "All swans are white"? (I did make a stupid choice here, sorry.)

I didn't say it should never be done, I said that it is always fallacious. There's a big difference between "fallacious argument" and "false conclusion".

Ah, I see. Now you are advocating fallacious reasoning when it works out. For instance, it is totally okay for math students to cancel the sixes in the numerator and denominator and conclude

16/64 = 1/4,

because it works, don't it?

I think we're done here. You have an addled misunderstanding of basic critical thinking and it's clear we're not getting closer to fixing that.

 

[caveman1917]

Then you're very confused, because appeal to authority is not a fallacy in deductive logic, since it occurs only in the context of inductive arguments.

That's peculiar, I just gave you a deductive argument which was invalid because of employing an appeal to authority.

Copi's definition will do: a type of incorrect argument.

Define "argument" and define "correct".

Oh? What about Copi's text? Did I type that for nothing?

Not for nothing, it's an opinion, but it doesn't define any of its terms so it's not very useful.

Why not look up WP's article on Appeal to Authority and see that it is commonly recognized that not every appeal to authority is fallacious?

Many things are "commonly recognized", that doesn't mean much by itself.

Or do you just know better than logicians like Copi or those cited in WP?

Probably, though that strongly depends on how they will choose to define their terms.

Strange, I've taught informal logic many times as part of a critical thinking class. But go on, keep making such silly pronouncements.

You taught guesswork and called it "logic"? Define "logic".

 

[Delphic Oracle]

No, appealing to reliable, trustworthy authorities is not a fallacy. You are simply wrong.

I could, of course, cite a half-dozen text books confirming my view, but you would call this appealing to authority (wrongly).

How do you intend to support your view? Don't get me wrong, repeating "yes it is" is a novel form of debate, not seen since The Argument Clinic, but it doesn't actually persuade.

Did Hannibal exist? What evidence (aside from the testimony of reliable sources, both contemporaneous and recent) could you provide to establish that he did? Or do you prefer to be in the dark on this matter?

The key difference here is independent sources who offer premises towards a conclusion and can be tested against each other, tested against physical evidence (archeological discoveries proving places Hannibal is purported to have been did exist at the time period in question). Corroboration is a valuable tool in assessing legal-historical arguments (where repeatable testing is not an available form of support).

Sometimes the other evidence can reveal how a source might be correct through a certain lens of interpretation that a plain reading would be misleading to take at face value. Homer's epics have truth to them and there's been remarkable increases in physical and recorded evidence to support events that were portrayed in them. In another example is the proposal that Moses parting the Red Sea really did happen, only it took place more as a low tide revealing a long spar to escape. Then some horrible timing for the pursuing Egyptian army who's heavy equipment got them mired in the muck of the spar and killed when the tide rolled back in. Take away the magic of silhouetted marine mammals backlit by moonglow through a wall of water and it suddenly seems a bit more plausible.

That is not the case with the claims about Russia for those of us with no security clearance. We have merely been given the names of some people who hold certain sensitive governmental posts. Actually a lot of the broadest claims come from 'unnamed senior intelligence officials' if we want to be fair about it.

So really we're at taking the word of an authority who claims that an unnamed authority claims that some evidence they have but won't show says their conclusion is correct.

Also, yes the historical record of reliability of many of the agencies in question are an issue for me. I do not mean I dismiss or take the opposite positions they do in all cases, I only mean they do not carry enough weight to base conclusions on. They have value in being evaluated as part of a broader body of evidence for testing against each other (as described in the Hannibal example) or potentially interpreting from the kind of lens they have shown a tendency to see things through.

Hell, there's almost no media publications I read without a sense of their tendencies and an attempt to both read it plain and 'decode' the bias.

 

[phiwum]

The key difference here is independent sources who offer premises towards a conclusion and can be tested against each other, tested against physical evidence (archeological discoveries proving places Hannibal is purported to have been did exist at the time period in question). Corroboration is a valuable tool in assessing legal-historical arguments (where repeatable testing is not an available form of support).

Sure, all of these considerations are relevant in considering whether an appeal to authority is strong. The argument in favor of Hannibal's existence has to be stronger than the argument regarding Russian involvement.

I nonetheless find the allegations more probable than not, but reasonable people can disagree.

 

[caveman1917]

No, this definition reduces it to purely deductive arguments once again. You ignore the fact that much of the time we use inductive arguments, and rightly so.

No, absolutely no. There are plenty of deductive frameworks available when classical logic won't suffice, including Bayesian inference or any number of alternatives. There is no need to use heuristics and guesswork, and certainly not to then call it "logic".

We appeal to analogy, for instance, or we draw conclusions about white swans of the sort already mentioned. Most of our informal fallacies are about bad inductive arguments, arguments whose premises do not suffice to ensure that the conclusion is probable.

Define "probable". As far as I know the term only makes sense in saying that an event in a probability space has a probability which is above some cut-off which we define as "probable". You have not given me a probability space hence you have not provided the context in which your claim about a conclusion being "probable" can be understood.

Only in the limit, but not at any particular finite number of observations.

Always. Likelihoods are always independent of priors. Are you sure you are not confusing posteriors with likelihoods?

And do it again with the negation of "All swans are white"? (I did make a stupid choice here, sorry.)

The calculation is trivial but long-winded, why are you asking this? Let N be the total number of swans, let N_black be the number of black ones. Then for each possible N_black from 0 to N, calculate the likelihood of seeing 275 white ones, and at the end combine them all to get an "All white" vs "Not all white" likelihood.

Ah, I see. Now you are advocating fallacious reasoning when it works out. For instance, it is totally okay for math students to cancel the sixes in the numerator and denominator and conclude

16/64 = 1/4,

because it works, don't it?

Wut?

I think we're done here. You have an addled misunderstanding of basic critical thinking and it's clear we're not getting closer to fixing that.

Maybe you should try defining your terms and giving arguments.

 

[caveman1917]

Yeah, you can't prove a negative

Of course we can prove a negative, we prove negatives all the time.

"There does not exist an even natural number strictly greater than 2 which is prime." Oh no! What are we to do? It's a negative, we'll never be able to prove it!

And even more generally, in any two-valued logic which includes elimination of double negation (which includes anything we'll ever use around here) there is absolutely no difference between proving a "negative" and proving a "positive". This entire "you can't prove a negative" thing is badly confused.

 

[Noztradamus]

Who has suggested that Russia tampered with the vote tallying?

That would be Senator Harry Reid(D) Nevada

http://www.nytimes.com/2016/08/30/us/politics/harry-reid-russia-tampering-election-fbi.html?_r=0

 

[Upchurch]

That would be Senator Harry Reid(D) Nevada

http://www.nytimes.com/2016/08/30/us/politics/harry-reid-russia-tampering-election-fbi.html?_r=0

Thank you.

That didn't seem very hard.

 

[caveman1917]

phiwum, I've noticed there are perhaps a couple of misunderstandings:

When I said "there is no such thing as informal logic" I didn't mean that the field classified under "informal logic" doesn't exist but that it is not a branch of logic, and calling it logic is a misappropriation of the term. It should really be called something like "educated guesswork" which covers it much more accurately than pretending it to be a form of logic.

When I said "who cares how we reason? We reason incorrectly all the time" I mean that a description of how people reason does not mean that the reasoning is correct. For example people reason by using confirmation bias, but just because that is "how we reason" doesn't mean an argument based on confirmation bias is correct.

 

[phiwum]

phiwum, I've noticed there are perhaps a couple of misunderstandings:

When I said "there is no such thing as informal logic" I didn't mean that the field classified under "informal logic" doesn't exist but that it is not a branch of logic, and calling it logic is a misappropriation of the term. It should really be called something like "educated guesswork" which covers it much more accurately than pretending it to be a form of logic.

When I said "who cares how we reason? We reason incorrectly all the time" I mean that a description of how people reason does not mean that the reasoning is correct. For example people reason by using confirmation bias, but just because that is "how we reason" doesn't mean an argument based on confirmation bias is correct.

If you prefer to call it "critical thinking" rather than "informal logic", I have no complaints. Historically, of course, these topics have been a part of logic, though these days, when people think of logic, they think of the formal stuff.

I disagree that it is mere educated guesswork. I find the view that inductive reasoning must be done in a Bayesian setting[1] to be a fantastic fiction. On the contrary, it's appropriate to reason inductively, through analogy and also to appeal to trustworthy, informed authorities. This kind of thinking is what we do everyday. When my car makes a funny noise and I recall that this was a symptom of a bad starter last time I heard it, then I am doing informal inductive reasoning, and quite appropriately, too.

You're right that we tend to reason poorly, but informal logic is prescriptive. It teaches one how to reason better, how to avoid fallacies and judge the relative strength of different informal arguments. That is the point of studying this subject. Very few students will do any serious technical work, but we hope that teaching them basic concepts of critical thinking will be useful to them on a daily basis.[2]

[1] I did bungle the term "likelihood" earlier.

[2] In practice, they forget everything they learned right after the final, of course. Except for the fallacies, which they misuse on web forums for the rest of their lives. I'm not fooling myself.

 

[zorro99]

Republican National Committee Chairman Reince Priebus, Mr. Trump’s White House chief-of-staff designate, conceded on Sunday that Russia was responsible for efforts to meddle in the 2016 presidential election but said Russia had done so in previous elections and “this is nothing new.”

“The fact that this particular hack was perpetrated by Russian entities is something that no one is disputing,” Mr. Priebus said in an interview on CBS.

http://www.wsj.com/articles/divide-over-russia-deepens-in-congress-1483927381?mod=e2fb

 

[Upchurch]

You are plain wrong.

Tell me:

What is the moon made of? How do know? Have you been there yourself?

Do vaccines cause autism? Have you conducted studies to confirm the consensus?

Is Einstein's Theory of Special Relativity true? How would you know?


Now, what do all these things have in common with the US intelligence community's report?

 

[uke2se]

Republican National Committee Chairman Reince Priebus, Mr. Trump’s White House chief-of-staff designate, conceded on Sunday that Russia was responsible for efforts to meddle in the 2016 presidential election but said Russia had done so in previous elections and “this is nothing new.”

“The fact that this particular hack was perpetrated by Russian entities is something that no one is disputing,” Mr. Priebus said in an interview on CBS.

http://www.wsj.com/articles/divide-over-russia-deepens-in-congress-1483927381?mod=e2fb

He should probably tell Donald Trump.

Not that the moron would listen.

This reminds me quite a bit of AGW denial. The steps of denial are being followed closely. First, deny it completely. Second, admit that it happened, but claim it's natural/happens all the time.

The third step will be, sure it happened, and it's really bad, but there's nothing we can do about it now.