Ask Slashdot: Why Are Scientists Constantly Surprised By What They Discover?

Slashdot reader dryriver asks about "the sheer number of times scientists consider something to be 'scientifically impossible', are badly disproven by some kind of new finding or discovery a few years later, and then express 'surprise' that 'X is indeed possible'." If you do a Google News search for the keywords "scientists were surprised" or similar, a huge number of science-related news articles contains a passage about "scientists being surprised" by what they discovered. There seems to be a great disparity between the mindset of inventors -- who always try to MAKE new things become possible -- and the mindset of many scientists, who seem unable or unwilling to consider that what "science holds to be true today" may not turn out to be quite so true tomorrow.

Here's the question: Why do many scientists, having knowledge of the fact that surprises in science happen all the time, continually express "surprise" when they find something unusual? If surprises in scientific research are so common, why are scientists still "surprised" by "surprise findings"?
"The surprising stuff is what we hear about, and there has to be some reason why it is surprising," argues gurps_npc in response to the original submission. "A common answer is that current state of science thinks the surprising stuff was impossible."

"The whole premise is flawed," counters long-time reader Martin+S. "Natural skepticism is an essential component of science." And long-time reader UnknownSoldier supplies a one-word answer: "Ego."

But how would you answer the question? Share your best thoughts in the comments. Why are scientists constantly surprised by what they discover?

Comments

It's a trick to get your papers published

[invalid_user]

In many scientific fields, especially mathematics (which is of course not technically a science but that's not the point here so let's not argue about that), results are often not interesting unless they are "surprising". Hence the tendency to exxegerate things.

There are also the occasions when scientists are pessimistic about certain results, and when these turned out well, they become pleasantly surprised.

So are scientists lying when they say they are surprised? No, they are indeed surprised. However, the level of surprise is low. It's a figure of speech.

For us to be alarmed, we would have to be "shocked" and "in disbelief".

Journalists and headline editors, not scientists

[skoskav]

This is a common joke in the skeptical community. To quote Steven Novella, paraphrasing Christopher Hitchens:

Journalists tend to have a limited pallet of story themes from which they choose, and then they conform the story to the chosen theme. Stories always need to be about something, such as corporate greed or government malfeasance, so that is the story that is told – regardless of the pesky facts.

Bad science journalism works that way also. That is why we can joke about common cliches, such as “Missing Link Discovered,” “Scientists Baffled,” and “It turns out everything we thought we knew was wrong.”

Clarke's First Law

[mentil]

"When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong." - Arthur C. Clarke's first law

If you really care to know...

[The Real Dr John]

As a neuroscientist I can tell you why many scientists in the life sciences are surprised by findings: shocker! It's because living systems are so absurdly complicated. Just take a look at what is known currently for any major biochemical pathway, or gene regulation, or mitochondrial metabolism, or protein trafficking in the cell. The complexity is mind boggling. Anyone who thinks you can wade into that abyss of unknowns with certainty hasn't done any biological research.

That's what we call selection bias.

[Sique]

All the expected results don't stir so much attention, as they were expected anyway. But as we know, the most exciting phrase in Science is not "Heureka!", it's "Well, that's odd." (often attributed to Isaac Asimov).

Things going according to plan don't make for exciting news. Discoveries that were planned for don't make for exciting news. Only the unexpected gets attention. If you find something you were expecting anyway, then there is nothing to be excited about.

You could even cite Claude Shannon: Information is the inverse of probability. If the Improbable happens, you get much more information than from an event highly probable. Thus yes, important discoveries are often not expected.

Re:Journalists and headline editors, not scientist

[Anonymous Coward]

For further details on bad journalism, read http://phdcomics.com/comics/archive.php?comicid=1174

It's amazing they aren't constantly surprised.

[Mal-2]

The universe is under no obligation to make sense to primates that evolved for life on the savannahs of Africa.
It's rather amazing that natural laws are amenable to logic, mathematics, and thought experiments, and that scientists so often guess right.
In other words, this is the wrong question. The question should be "Why is the natural world predictable in such detail, and why are we getting it right more often than not?"

Different meaning

[pz]

Scientists aren't surprised. That's just the language that is used to describe findings that don't match up with existing dogma.

As another poster suggested, the natural world is exceedingly complex. Physics aside, our models of it tend to be simple. Simple models perforce do not capture complexity, and thus, are often wrong when you test them beyond their domain.

If you, as a scientist, aren't constantly stumbling across unexpected results (which are written as surprises, that term has a different meaning in scientific papers than in the general public), then you aren't exploring new areas. As a scientist, you work by taking an existing model or hypothesis, and pushing it to its limits, finding where it breaks down, and creating a new, better model that accommodates a wider area. There are precious few cases where such models are sufficiently complete that we have run out of things to test ... low-energy fundamental particle physics seems to be the best-known one. In biology, which is the field I work in, we aren't even remotely close.

Take paleontology, for instance. One a seemingly monthly basis, new dinosaur species are being discovered, or old bones are realized to have been put together wrong, or new details about extinctions have been discovered. For that field, much of the surprise comes from additional data sources -- our older, simpler models were based on less data, and with additional information, better models can be built. Dinosaurs, when I was a kid, were thought to all be lizard-like in appearance. Recent discoveries of exceedingly well-preserved specimens suggest many of them had feathers, and were colored.

Take planetary sciences / cosmology. We have discovered a vast trove of objects in our solar system, thanks to new streams of data. We have discovered large numbers of planets beyond our solar system, also thanks to new streams of data. The better we build our telescopes and sensors, the better a picture of the cosmos we get. Each increase in available resolution continues to bring surprises because we do not have fully-developed models of the universe.

Take geology. Plate tectonics was validated only about 50 years ago. We don't know for sure that the same thing happens on other planets.

And biology. The combination of Darwin, and Watson and Crick seemed to explain all of evolution. Except that, as we look more and more closely, there *are* acquired traits that are inherited ... they're just not the dominant means of evolution. Our tools are getting better and measuring with finer molecular detail, revealing secrets of the scaffolding around DNA and the immense role it plays in determining externally observable characteristics.

Or sleep. We actually understand much of the metabolic mechanism for sleep, now. There is a real rejuvenation process. But we wouldn't have understood that without new tools that allow us to probe at high temporal and spatial resolution, and with fine molecular resolution using genetic tools.

In short, scientists are surprised because we discover new things all the time. We remain on the cusp of wide troves of knowledge, all of which is new. Each new revolution in data collection brings with it a new, unexplored realm and, as is written in many papers, surprises.

Unexpected results

[gordona]

I studied some of the mechanical and contractile properties of smooth muscle. In spite of vast morphological differences between smooth muscle and striated (skeletal) muscle, smooth muscle demonstrated qualitatively similar results as striated muscle. https://www.pnas.org/content/7.... The surprise here is that form and function do not necessarily follow each other.

Re: Because it gives you more funding

[wooferhound]

The Scientests aren't actually surprised
It's the Reporters writing the news stories that are surprised

Re: Because it gives you more funding

[tsa]

Indeed, that was what I came up with first too. I was a scientist for a long time, and of course sometimes nature surprises you, but to get funding you need to use every superlative in your tool set, and 'suprising' seems to work well even with stuff you didn't find that surprising. And ince you have funding the money givers will want to hear great stories, so there we go again.

Surprising results often wrong

[Roger W Moore]

They didn't act surprised about the Higgs boson, because it was hot enough on its own.

No, we were not surprised because it had been predicted as a solution in 1964. Then we built a massive collider and two huge experiments specifically to search for it. Anyone who expressed surprise at finding it in 2012 would have to have been an idiot.

Indeed the vast majority of recent surprises in particle physics have been exactly the opposite to what the article suggests. In our case, the surprises have generally turned out to be someone making a serious error. For example, the claim of a faster than light neutrino surprised everyone because it violated relativity. The eventual result was that it was caused by a cable that was not properly plugged in, which was a result that surprised nobody.

A similar thing happened a few years ago at the LHC where both experiments started to see signs of a surprising new resonance. However, as more data were collected the significance declined and it appears that it was just a statistical fluke. So in my experience surprising results are usually the ones that turn out to be wrong which is what you expect when you have a good understanding of what you are studying.

If you have lots of surprising results which turn out to be right then you clearly have a very poor understanding of whatever you are studying because the predictions of your theoretical model are constantly being proved wrong.

Re:God continuously invents science.

[PsychoSlashDot]

You should zip up your pants; your agenda is showing.

What is interesting about science it that science - at least today - KNOWS very well that there is A LOT that we have yet to explain fully or discover. Science KNOWS that we humans, basically, know only how SOME of how the universe we live in functions. And yet many scientists are SO CERTAIN that there is no God, or any kind sentient intelligence that created or designed the vast universe that we are a tiny part of.

Your language is sloppy, and it suggests your analysis may be as well. Yes, science accepts that it is not yet "complete". There are explanations for a few observed phenomenon that are not yet incorporated into the existing body of scientific understanding. Introduction of a God or multiple Gods into the discussion is pretty much irrelevant to the "completeness" of scientific understanding. Why? Because statistics. So far, zero of observed phenomenon that have been explained have required the involvement of a God or multiple gods. Zero. None. Nada. Zilch. Bupkiss.

More, the obscurity of the few observed phenomenon that have not yet been incorporated into scientific understanding continues to become increasingly massive. Invocation of God or Gods used to be required to "explain" such trivial experiences as fire, disease, earthquakes, lights in the sky, and pregnancy. Now we understand these things, to such a degree that God or Gods are no longer required for any of them. We now live in a time where "don't share needles" is all the wisdom required, and "go ahead and share needles with another junkie but you'll be fine as long as you pray" is laughable. It's comedic. Even among the religious community, reliance on scientific understanding is widespread enough that they would view someone who just prays they don't contract AIDS from unprotected sex with a carrier as delusional.

To recap, the utter and total lack of requirement for God or Gods in 100% of what we know - which is vast - makes the lack of belief in God quite understandable. And mostly, sensible.

This is not just contradictory, but downright dangerous.

You're going to have to demonstrate that. There's no contradiction. At all. "I don't know everything, but nothing I do know - which is virtually the entire scope of my observed experience - even remotely suggests there is a God or Gods, so I suspect there is no such entity or entities." Not contradictory. Or dangerous.

Basically, scientists who know VERY WELL that they only UNDERSTAND PART OF FUNCTIONING THE UNIVERSE and HAVE NO IDEA WHATSOEVER WHERE OUR UNIVERSE CAME FROM are ABSOLUTELY CERTAIN that there can be no such thing as God.

Oh. You don't understand atheists or scientists. Let me help. First, yes, atheists believe there are no Gods. But... were there evidence or - Heaven forbid - proof of the existence of such, they'd be willing to accept they had been mistaken. Second, scientists are even more willing to accept change. That's how science works. If a theory or working model of a situation is tested and demonstrated by further data to be false, the theory or working model is either invalidated or updated to incorporate the new data. Neither atheists nor scientists are - as a rule - certain they are right. The believe, according to the evidence at hand, that they are. But certainty is not part of their worldview.

WTF? That's about as logical as saying "I have never actually physically travelled to Ethiopia, but I know everything everything there is to know about Ethiopa nevertheless."

False. It's exactly as logical as saying "to date virtually everything humanity has observed has had a non-deity explanation and every day more of the incredibly obscure observations we haven't explained are explained, and continue to have non-deity explanations, so the unanimous body of evidence predicts the non-existence of deities."

Eh .... no.

[Freischutz]

The Scientests aren't actually surprised It's the Reporters writing the news stories that are surprised

No, scientists do not operate in absolute certainty of what they will discover, they are regularly surprised by what they discover. It is the religionists who have absolute certainty because they are the only ones I have met that claim they can explain everything in the universe, ... with a collection of ancient religious texts and the fickle opinions of their clergy.

Re:Eh .... no.

[ceoyoyo]

It you expected it, it's not much of a discovery, is it?

Re:Remember: Clarke's First Law is fiction

[BorisAmmerlaan]

According to Shirlock Holmes "when you have eliminated the possible, only the impossible remains!"

I do not know who Shirlock Holmes is, but the actual Sherlock Holmes quote is: "when you have eliminated the impossible, whatever remains, however improbable, must be the truth"

So... exactly the opposite of what you said.