Sometimes to stop arguing with someone, we say: “You have a right to your opinion.”
But is that true? May we believe whatever we want to believe?
In his essay “The Ethics of Belief,” British mathematician and philosopher W.K. Clifford (1845–1879) argues that the answer is “no.” He claims that “it is wrong always, everywhere, and for anyone, to believe anything upon insufficient evidence.”
I think the article is about sufficient evidence, not certainty. Using your example, you can have sufficient evidence for things that never happened before, no? Isn't that the point of training...
I think the article is about sufficient evidence, not certainty. Using your example, you can have sufficient evidence for things that never happened before, no? Isn't that the point of training regimen, sports medicine, kinesiology, etc? 🤔
The first person who trained to run a sub 4-minute mile was defying all sufficient evidence. So was the first person to build an Airplane, or invent the Lightbulb. Their dissident belief is what...
The first person who trained to run a sub 4-minute mile was defying all sufficient evidence.
So was the first person to build an Airplane, or invent the Lightbulb.
Their dissident belief is what drove them to succeed.
The 'evidence' changes each time someone transcends the 'impossible'.
That is a cogent perspective. I have a different view. I believe that when someone invent something like a lamp, they generally do so while supported by a bunch of data (partly accumulated through...
That is a cogent perspective. I have a different view.
I believe that when someone invent something like a lamp, they generally do so while supported by a bunch of data (partly accumulated through many previous attempts) which is sufficient evidence that a lamp is at least possible enough to warrant their effort.
But maybe you're thinking more on the idea than execution? Cause yeah, for some guy to wake up and think "I'm gonna make an electric lamp!", I don't think any evidence is required!
To actually make the lamp, I think at least some evidence is unavoidable.
I agree. The inventor, alone, feels they have sufficient evidence to try. And I agree, sufficient evidence is defined by the individual. My beef with this whole theory is: I don’t think a person...
I agree. The inventor, alone, feels they have sufficient evidence to try.
And I agree, sufficient evidence is defined by the individual.
My beef with this whole theory is:
I don’t think a person can truly Believe something, if they are sufficiently convinced against it.
I don’t think that is possible.
Can you think of an example? (a true inner-Belief, despite convincing counter-evidence)
At the risk of going meta, I don’t think you can possibly have sufficient evidenve to hold that belief. Especially w.r.t. technology, these achievements did not happen in a vacuum, not even...
At the risk of going meta, I don’t think you can possibly have sufficient evidenve to hold that belief.
Especially w.r.t. technology, these achievements did not happen in a vacuum, not even weirdly divergent efforts or understandings. In both the cases you cited, multiple teams around the world were trying to achieve these milestones and the physics underlying their engineering efforts was reasonably well understood. Everyone contemporaneous believed it would only be a matter of time, or more precisely a matter of sufficient tuning, before these were achieved.
I’m saying... They were brave to study flight - and risk their lives to try - based on theories held by a small group of people. ...And when everyone else had failed or died <== (pretty hefty...
I’m saying...
They were brave to study flight - and risk their lives to try - based on theories held by a small group of people.
...And when everyone else had failed or died <== (pretty hefty evidence to consider when calculating 'sufficient evidence').
To me - their belief far exceeded the evidence.
Thus, it is not 'wrong for anyone, to believe anything upon insufficient evidence.'
So you're saying...if I time-travel to 1890 (before human flight)....1000 out of 1000 people would say airplanes were guaranteed to exist? Even when previous attempts had resulted in death? Source?
Everyone contemporaneous believed...
So you're saying...if I time-travel to 1890 (before human flight)....1000 out of 1000 people would say airplanes were guaranteed to exist?
Even when previous attempts had resulted in death?
Obviously i was not saying that every individual alive at that time, that’s preposterous. Also, this is a casual conversation, and also tildes. I neither expect nor promote “sources”.
Obviously i was not saying that every individual alive at that time, that’s preposterous.
Also, this is a casual conversation, and also tildes. I neither expect nor promote “sources”.
Yeah...the ‘source’ request was a joke mate (of course you can’t look-up what ‘everyone believed’ in 1890). That is why...in your words...“I don’t think you can possibly have sufficient evidence...
Yeah...the ‘source’ request was a joke mate (of course you can’t look-up what ‘everyone believed’ in 1890).
That is why...in your words...“I don’t think you can possibly have sufficient evidence to hold that belief.”
Evidence doesn't work that way. Since it actually is possible, there obviously (in hindsight) wasn't enough evidence to rule it out completely. The people who believed it was impossible were far...
Evidence doesn't work that way.
Since it actually is possible, there obviously (in hindsight) wasn't enough evidence to rule it out completely. The people who believed it was impossible were far too certain in their beliefs.
There may have enough evidence to believe that it's hard to do, or unlikely for any particular runner to do it, taking an outside view. But proving something impossible is a lot harder than that. They should have hedged their bet.
But for someone who hedged, it's more difficult to decide if they got the probabilities wrong. Let's say someone said there's a 1 in a hundred chance of someone breaking the four minute mile this year, and then someone does it. Were they wrong? That's not something you can judge from a single prediction - you have to look at their overall track record.
Right. That’s why I put the word ‘impossible’ in ‘quotes’... Bannister’s Belief eclipsed the opinions of closed-minded ‘experts’, who wrongly called it ‘impossible’. The evidence - at the time -...
Right. That’s why I put the word ‘impossible’ in ‘quotes’...
Bannister’s Belief eclipsed the opinions of closed-minded ‘experts’, who wrongly called it ‘impossible’.
The evidence - at the time - was:
Every prior human attempt had failed, including his own. (<== that was the data).
Previous failures are statistical evidence, but statistical evidence alone doesn't prove impossibility. Obviously this proves too much because it says there's never a first time for anything....
Previous failures are statistical evidence, but statistical evidence alone doesn't prove impossibility. Obviously this proves too much because it says there's never a first time for anything. Context matters.
Opinions of doctors and scientists may be indirect evidence to the extent that they are evidence-based. But you'll need to understand the evidence that they base their opinions on to understand how it can go wrong.
This is often too much work, so it's usually a reasonable shortcut to trust expert opinions, but when you really care you want to dig into the details.
This is why I hate the meme that "doing your own research" is somehow wrong. Why do high schools and colleges teach students to do research if it's a bad thing?
But it's certainly easy to screw it up. For doing research to be educational you need to be curious enough to want to learn things, skeptical enough to question what you read, and humble enough to keep in mind that you might nonetheless be wrong.
Also, sometimes to make real progress, you need to be persistent enough to keep looking.
You are right - doing your own research is a very good thing. I agree, future outcomes can absolutely outshine past results. It is good that you don't blindly trust 'experts'...who knows what...
You are right - doing your own research is a very good thing.
I agree, future outcomes can absolutely outshine past results.
It is good that you don't blindly trust 'experts'...who knows what their motives are?
I enjoy your curiosity and careful thinking - those are great traits. Keep it up!
Also, I thought of an interesting question: do you believe that a three minute mile is impossible? What evidence do you have and what odds would you give? Do you think the power of belief has...
Also, I thought of an interesting question: do you believe that a three minute mile is impossible? What evidence do you have and what odds would you give?
Do you think the power of belief has anything to do with it?
Sure, a 3-minute mile is possible. The fastest 400m dash is 43.03 seconds. One mile is 4 x 400m. Theoretically, someone could do it already (if they could keep that 400m pace) ...'experts' will...
Sure, a 3-minute mile is possible.
The fastest 400m dash is 43.03 seconds.
One mile is 4 x 400m.
4 * 43.04 = 172 seconds = 2 minutes 52 seconds
Theoretically, someone could do it already (if they could keep that 400m pace)
...'experts' will say..."well there must be some limit"...
Sure...unless...
What if someone develops a superior running technique?
What if humans grow to be 9 feet tall, and cover 40% more distance with each stride?
Or -- What if humans blossom some dormant part of the mind, allowing us to teleport (like the quantum-physics theory of a particle existing in two places at once)...
Here's the way I would approach it: Suppose someone is trying to run a three minute mile and they want me to give them $10,000 if they do it. If I truly think it's impossible then I should say...
Here's the way I would approach it:
Suppose someone is trying to run a three minute mile and they want me to give them $10,000 if they do it.
If I truly think it's impossible then I should say "sure" because I'll never have to pay them. The current fastest time is 3:43 and it seems pretty unlikely that they'll ever win.
However, it's kind of a weird bet. Why are they making this bet? What do they know that I don't? Did I misunderstand something?
So, I don't think I'll agree to this for free. Maybe if they offered to pay me $100 I'll think about it? Maybe.
So that's 100 to 1 odds for something I think will never happen. I'm not sure I would ever make any bet giving someone any better odds than that, because it's not worth the risk that I screwed up. When it comes down to it I'm not really interested in taking bizarre financial risks for pitiful sums, because I don't trust myself not to screw up that much.
In finance this is sort of bet is called "picking up pennies in front of a steamroller" and I'm not into that sort of thing.
On the other hand, when buying insurance, I take tiny odds more seriously, at least for catastrophic risks. The risk of a house fire is probably much less than 1 in 100 but I would still buy fire insurance.
I don't buy insurance for every tiny risk though. Maybe I'm not entirely consistent about it.
Let's ask him... 4 Roger Bannister Quotes: "It is the brain, not the heart or lungs, that is the critical organ." "Doctors and scientists said breaking the four-minute mile was impossible, that...
Let's ask him...
4 Roger Bannister Quotes:
"It is the brain, not the heart or lungs, that is the critical organ."
"Doctors and scientists said breaking the four-minute mile was impossible, that one would die in the attempt. Thus, when I got up from the track after collapsing at the finish line, I figured I was dead."
"To move into the lead is an act requiring fierceness and confidence. But fear must play some part...no relaxation is possible, and all discretion is thrown into the wind."
"Without the concentration of the mind and the will, performance will not result."
I'd say he held a Very strong belief that running a 4-minute mile was possible.
So to believe in something, you must have evidence. But, doesn't evidence gained through research usually follow belief? That is, (almost) no one is out there trying to prove something they don't...
So to believe in something, you must have evidence.
But, doesn't evidence gained through research usually follow belief? That is, (almost) no one is out there trying to prove something they don't believe, unless it's to disprove said thing.
Also, I do have a right to my opinion, in as much as not everything can be proven or disproven. I cannot stand musicals, so it is my opinion that they are bad. This is highly subjective, and unable to be proven or disproven objectively.
Let's say you're sure you hate musicals and then end up going to one through social obligation and find yourself enjoying it despite everything. Then you were wrong to say that you could never...
Let's say you're sure you hate musicals and then end up going to one through social obligation and find yourself enjoying it despite everything.
Then you were wrong to say that you could never enjoy them and deprived yourself of a source of happiness.
But I think this is only a little wrong because there are other sources of happiness. We are wrong about things all the time and usually the consequences aren't all that big a deal.
I think the article is directionally right that having confident wrong beliefs often has bad consequences. But it's mistaken because there is no sense of scale. Consequences vary greatly. Most situations of being wrong have far lower consequences than a sunken ship. Often it matters so little that it's unknowable, lost in the noise.
There's also the problem that a partially wrong theory may still be useful, or it might be subtly wrong in a way that's disastrous, and we might not know which it is in advance.
I think in many situations it's morally wrong not to hedge your bets. It's a fact of life that ships do sink sometimes. Did they have enough lifeboats? The hypothetical example seems like a great argument for better disaster preparation.
Proper scientific research does not try to prove or disprove anything; it merely seeks to uncover the truth. A hypothesis may be presented as a starting point for an experiment or other line of...
Proper scientific research does not try to prove or disprove anything; it merely seeks to uncover the truth. A hypothesis may be presented as a starting point for an experiment or other line of scientific inquiry, but it is not a statement of belief.
Often however, a "hunch" or belief underlies research: You think X is the case, without evidence to support that (and absent evidence to the contrary) - purely on evidence, the jury is still out,...
Often however, a "hunch" or belief underlies research: You think X is the case, without evidence to support that (and absent evidence to the contrary) - purely on evidence, the jury is still out, but you have subjective reasons to think it's an interesting direction to pursue. You're absolutely going in trying to prove X. Is that the humanist ideal of scientific research? Maybe not, but I'd argue it doesn't take away from the scientific method - Science won't stop working because I'm biased in favor of my hypothesis. That's what peer review etc. is for.
E: This is of course, imo, not in conflict with the blog article: I investigate a certain topic because I have sufficient evidence to believe that it is novel and that truth can be found out. Novelty is one of my goals as a scientist. A guarantee of X working before I start the research is an unreasonably high bar and unnecessary. In that sense, I'm acting both in like with Clifford and in line with scientific theory.
There's also the difference between believing something and acting upon that. If I believe I know tomorrow's lottery number because I have some insider info... Let's say I put my odds at 1 in 100 of having the right numbers. It's perfectly rational to say "those odds are insufficient evidence to believe I've got the right numbers". It is however also perfectly rational to act on that insufficient evidence, because a 1 in 100 chance of a ROI of one million; that's great. Risk calculations are an entirely different thing separate from belief. Well, closely connected, but distinct.
Right, but you will never hear a reputable scientist go about proclaiming their hypothesis to be indisputable truth until they have the data to back it up. This is in contrast to say, a religious...
Right, but you will never hear a reputable scientist go about proclaiming their hypothesis to be indisputable truth until they have the data to back it up.
This is in contrast to say, a religious preacher, who will assert the truthfulness of their "hypothesis" that a certain book is of divine origin or that the Earth is 6,000 years old, without the backing of any empirical evidence. Or any number of other superstitious beliefs (psychics, homeopaths, "ghost hunters"). That is my interpretation of what the author means by it being "wrong" to believe without sufficient evidence.
Fair, fair. I think this whole thread is about "acting on insufficiently evidenced belief", when acting on beliefs is separate from holding beliefs based on insufficient evidence. I think that's a...
Fair, fair. I think this whole thread is about "acting on insufficiently evidenced belief", when acting on beliefs is separate from holding beliefs based on insufficient evidence. I think that's a crucial distinction. I don't need sufficient evidence that something is true or possible for me to try to prove it of achieve it. I would be foolish to try it in spite of sufficient evidence, but in the absence of evidence, there's opportunity... or risk, if you like that flavor better. And risk calculations factor in rewards of the outcomes: How famous will I be for running a sub-4-minute mile? How rich can I get with electric lighting? Those questions factor into my actions, but not my beliefs.
Something's been bugging me about the quote, and I think I figured it out. Always? For anyone? That's a mighty strong claim. What evidence is there for such an absolute claim? Oh, he had none, it...
Something's been bugging me about the quote, and I think I figured it out.
“it is wrong always, everywhere, and for anyone, to believe anything upon insufficient evidence.”
Always? For anyone? That's a mighty strong claim. What evidence is there for such an absolute claim? Oh, he had none, it was just something he believed, an opinion. All right then.
What about the first person to run a sub 4-minute mile?
He believed it was possible...against all evidence - was that morally wrong?
I think the article is about sufficient evidence, not certainty. Using your example, you can have sufficient evidence for things that never happened before, no? Isn't that the point of training regimen, sports medicine, kinesiology, etc? 🤔
The first person who trained to run a sub 4-minute mile was defying all sufficient evidence.
So was the first person to build an Airplane, or invent the Lightbulb.
Their dissident belief is what drove them to succeed.
The 'evidence' changes each time someone transcends the 'impossible'.
That is a cogent perspective. I have a different view.
I believe that when someone invent something like a lamp, they generally do so while supported by a bunch of data (partly accumulated through many previous attempts) which is sufficient evidence that a lamp is at least possible enough to warrant their effort.
But maybe you're thinking more on the idea than execution? Cause yeah, for some guy to wake up and think "I'm gonna make an electric lamp!", I don't think any evidence is required!
To actually make the lamp, I think at least some evidence is unavoidable.
I agree. The inventor, alone, feels they have sufficient evidence to try.
And I agree, sufficient evidence is defined by the individual.
My beef with this whole theory is:
I don’t think that is possible.
At the risk of going meta, I don’t think you can possibly have sufficient evidenve to hold that belief.
Especially w.r.t. technology, these achievements did not happen in a vacuum, not even weirdly divergent efforts or understandings. In both the cases you cited, multiple teams around the world were trying to achieve these milestones and the physics underlying their engineering efforts was reasonably well understood. Everyone contemporaneous believed it would only be a matter of time, or more precisely a matter of sufficient tuning, before these were achieved.
I’m saying...
They were brave to study flight - and risk their lives to try - based on theories held by a small group of people.
...And when everyone else had failed or died <== (pretty hefty evidence to consider when calculating 'sufficient evidence').
To me - their belief far exceeded the evidence.
Thus, it is not 'wrong for anyone, to believe anything upon insufficient evidence.'
...Sometimes that is what makes a person heroic.
So you're saying...if I time-travel to 1890 (before human flight)....1000 out of 1000 people would say airplanes were guaranteed to exist?
Even when previous attempts had resulted in death?
Source?
Obviously i was not saying that every individual alive at that time, that’s preposterous.
Also, this is a casual conversation, and also tildes. I neither expect nor promote “sources”.
Yeah...the ‘source’ request was a joke mate (of course you can’t look-up what ‘everyone believed’ in 1890).
That is why...in your words...“I don’t think you can possibly have sufficient evidence to hold that belief.”
Evidence doesn't work that way.
Since it actually is possible, there obviously (in hindsight) wasn't enough evidence to rule it out completely. The people who believed it was impossible were far too certain in their beliefs.
There may have enough evidence to believe that it's hard to do, or unlikely for any particular runner to do it, taking an outside view. But proving something impossible is a lot harder than that. They should have hedged their bet.
But for someone who hedged, it's more difficult to decide if they got the probabilities wrong. Let's say someone said there's a 1 in a hundred chance of someone breaking the four minute mile this year, and then someone does it. Were they wrong? That's not something you can judge from a single prediction - you have to look at their overall track record.
Right. That’s why I put the word ‘impossible’ in ‘quotes’...
The evidence - at the time - was:
Every prior human attempt had failed, including his own. (<== that was the data).
Doctors and scientists said he would die trying.
He did not agree that it was ‘impossible’.
His Belief was stronger than their evidence.
That is why he succeeded.
Previous failures are statistical evidence, but statistical evidence alone doesn't prove impossibility. Obviously this proves too much because it says there's never a first time for anything. Context matters.
Opinions of doctors and scientists may be indirect evidence to the extent that they are evidence-based. But you'll need to understand the evidence that they base their opinions on to understand how it can go wrong.
This is often too much work, so it's usually a reasonable shortcut to trust expert opinions, but when you really care you want to dig into the details.
This is why I hate the meme that "doing your own research" is somehow wrong. Why do high schools and colleges teach students to do research if it's a bad thing?
But it's certainly easy to screw it up. For doing research to be educational you need to be curious enough to want to learn things, skeptical enough to question what you read, and humble enough to keep in mind that you might nonetheless be wrong.
Also, sometimes to make real progress, you need to be persistent enough to keep looking.
You are right - doing your own research is a very good thing.
I agree, future outcomes can absolutely outshine past results.
It is good that you don't blindly trust 'experts'...who knows what their motives are?
I enjoy your curiosity and careful thinking - those are great traits. Keep it up!
Also, I thought of an interesting question: do you believe that a three minute mile is impossible? What evidence do you have and what odds would you give?
Do you think the power of belief has anything to do with it?
Sure, a 3-minute mile is possible.
The fastest 400m dash is 43.03 seconds.
One mile is 4 x 400m.
Theoretically, someone could do it already (if they could keep that 400m pace)
...'experts' will say..."well there must be some limit"...
Sure...unless...
What if someone develops a superior running technique?
What if humans grow to be 9 feet tall, and cover 40% more distance with each stride?
Or -- What if humans blossom some dormant part of the mind, allowing us to teleport (like the quantum-physics theory of a particle existing in two places at once)...
....Then you’re lookin at a zero-second-mile.
Here's the way I would approach it:
Suppose someone is trying to run a three minute mile and they want me to give them $10,000 if they do it.
If I truly think it's impossible then I should say "sure" because I'll never have to pay them. The current fastest time is 3:43 and it seems pretty unlikely that they'll ever win.
However, it's kind of a weird bet. Why are they making this bet? What do they know that I don't? Did I misunderstand something?
So, I don't think I'll agree to this for free. Maybe if they offered to pay me $100 I'll think about it? Maybe.
So that's 100 to 1 odds for something I think will never happen. I'm not sure I would ever make any bet giving someone any better odds than that, because it's not worth the risk that I screwed up. When it comes down to it I'm not really interested in taking bizarre financial risks for pitiful sums, because I don't trust myself not to screw up that much.
In finance this is sort of bet is called "picking up pennies in front of a steamroller" and I'm not into that sort of thing.
On the other hand, when buying insurance, I take tiny odds more seriously, at least for catastrophic risks. The risk of a house fire is probably much less than 1 in 100 but I would still buy fire insurance.
I don't buy insurance for every tiny risk though. Maybe I'm not entirely consistent about it.
Did Roger Bannister hold an immutable belief that running a 4-minute mile was possible, or was it merely a hypothesis he set about testing?
Let's ask him...
4 Roger Bannister Quotes:
"It is the brain, not the heart or lungs, that is the critical organ."
"Doctors and scientists said breaking the four-minute mile was impossible, that one would die in the attempt. Thus, when I got up from the track after collapsing at the finish line, I figured I was dead."
"To move into the lead is an act requiring fierceness and confidence. But fear must play some part...no relaxation is possible, and all discretion is thrown into the wind."
"Without the concentration of the mind and the will, performance will not result."
I'd say he held a Very strong belief that running a 4-minute mile was possible.
So to believe in something, you must have evidence.
But, doesn't evidence gained through research usually follow belief? That is, (almost) no one is out there trying to prove something they don't believe, unless it's to disprove said thing.
Also, I do have a right to my opinion, in as much as not everything can be proven or disproven. I cannot stand musicals, so it is my opinion that they are bad. This is highly subjective, and unable to be proven or disproven objectively.
Let's say you're sure you hate musicals and then end up going to one through social obligation and find yourself enjoying it despite everything.
Then you were wrong to say that you could never enjoy them and deprived yourself of a source of happiness.
But I think this is only a little wrong because there are other sources of happiness. We are wrong about things all the time and usually the consequences aren't all that big a deal.
I think the article is directionally right that having confident wrong beliefs often has bad consequences. But it's mistaken because there is no sense of scale. Consequences vary greatly. Most situations of being wrong have far lower consequences than a sunken ship. Often it matters so little that it's unknowable, lost in the noise.
There's also the problem that a partially wrong theory may still be useful, or it might be subtly wrong in a way that's disastrous, and we might not know which it is in advance.
I think in many situations it's morally wrong not to hedge your bets. It's a fact of life that ships do sink sometimes. Did they have enough lifeboats? The hypothetical example seems like a great argument for better disaster preparation.
Proper scientific research does not try to prove or disprove anything; it merely seeks to uncover the truth. A hypothesis may be presented as a starting point for an experiment or other line of scientific inquiry, but it is not a statement of belief.
Often however, a "hunch" or belief underlies research: You think X is the case, without evidence to support that (and absent evidence to the contrary) - purely on evidence, the jury is still out, but you have subjective reasons to think it's an interesting direction to pursue. You're absolutely going in trying to prove X. Is that the humanist ideal of scientific research? Maybe not, but I'd argue it doesn't take away from the scientific method - Science won't stop working because I'm biased in favor of my hypothesis. That's what peer review etc. is for.
E: This is of course, imo, not in conflict with the blog article: I investigate a certain topic because I have sufficient evidence to believe that it is novel and that truth can be found out. Novelty is one of my goals as a scientist. A guarantee of X working before I start the research is an unreasonably high bar and unnecessary. In that sense, I'm acting both in like with Clifford and in line with scientific theory.
There's also the difference between believing something and acting upon that. If I believe I know tomorrow's lottery number because I have some insider info... Let's say I put my odds at 1 in 100 of having the right numbers. It's perfectly rational to say "those odds are insufficient evidence to believe I've got the right numbers". It is however also perfectly rational to act on that insufficient evidence, because a 1 in 100 chance of a ROI of one million; that's great. Risk calculations are an entirely different thing separate from belief. Well, closely connected, but distinct.
Right, but you will never hear a reputable scientist go about proclaiming their hypothesis to be indisputable truth until they have the data to back it up.
This is in contrast to say, a religious preacher, who will assert the truthfulness of their "hypothesis" that a certain book is of divine origin or that the Earth is 6,000 years old, without the backing of any empirical evidence. Or any number of other superstitious beliefs (psychics, homeopaths, "ghost hunters"). That is my interpretation of what the author means by it being "wrong" to believe without sufficient evidence.
Fair, fair. I think this whole thread is about "acting on insufficiently evidenced belief", when acting on beliefs is separate from holding beliefs based on insufficient evidence. I think that's a crucial distinction. I don't need sufficient evidence that something is true or possible for me to try to prove it of achieve it. I would be foolish to try it in spite of sufficient evidence, but in the absence of evidence, there's opportunity... or risk, if you like that flavor better. And risk calculations factor in rewards of the outcomes: How famous will I be for running a sub-4-minute mile? How rich can I get with electric lighting? Those questions factor into my actions, but not my beliefs.
Something's been bugging me about the quote, and I think I figured it out.
Always? For anyone? That's a mighty strong claim. What evidence is there for such an absolute claim? Oh, he had none, it was just something he believed, an opinion. All right then.