As society advances, technological discovery adds more and more to our accumulated knowledge. As such, scientists need to learn more and more to get to the point where they start exploring the boundaries of our knowledge. Sir Isaac Newton for example never had to take Calculus in college.

Certainly it is easier to learn things than discover them, and things like Calculus made a bunch of other things easier. But it seems as though there will be ever increasing prior work in a field. And it seems that even though some advances will make things less severely complicated (or at least better abstracted), the overall trend of discovery will be for more and more complex things.

Won't this eventually overtake people's lifespans? What would it take to create that scenario where becoming skilled enough in a field to advance it leaves no time to actually advance it?

  • $\begingroup$ Related to my answer: are you talking about Discovery as "the formation of new novel ideas to be tested" or "the testing and acceptance of those ideas as truth by the general population?" $\endgroup$ – Cort Ammon Dec 10 '14 at 16:21
  • 2
    $\begingroup$ @CortAmmon - The advancement of scientific knowledge. Sometimes that means verifying theories, sometimes it means developing new processes, sometimes it means creating better theories. It's fairly uncommon even today for the general population to know/accept discoveries. $\endgroup$ – Telastyn Dec 10 '14 at 16:29
  • 1
    $\begingroup$ One thing to note, techniques or technologies to make learning much faster could help this a lot. It is also noting the constant increase in useful lifespan. $\endgroup$ – Vality Dec 11 '14 at 3:26
  • $\begingroup$ I'm reminded of a certain scene from Star Trek: The Next Generation, in which a small child (elementary school age) was crying and complaining to his mother about having to go to school and take Calculus. It implied that this is the future's solution to this problem: learn more, earlier in life. $\endgroup$ – Mason Wheeler Dec 13 '14 at 19:45
  • $\begingroup$ This situation reminds me of the Startide Rising by David Brin (and the rest of the series). The uplift universe reaches a point where just about everything of importance has been discovered and optimized already. Why spend the time designing a spaceship, if you can look up the plans at the library for a nearly optimal design. Entire species spend their time reading and cross-referencing the library, since that is the fastest way to improve their power. In an arms race, they find there is no time for independent discovery or re-inventing the wheel. $\endgroup$ – Xantix Nov 16 '17 at 22:04

20 Answers 20


It has already happened!

In 322 BCE, Aristotle died. He may be considered one of the last people in the world who knew everything about everything at the time.

Since then, advances have been made but very few people span more than one discipline (chemistry, physics, biology); nowadays, it is remarkable when a person spans more than a single sub discipline.

But more advances are being made today than ever before, not because of great individuals who know everything but by the societal structure. Advances in computing, for example, are being incorporated into everybody's lives allowing for further advances without the people using the computers knowing everything (or even anything) about those computers.

Though one thing you might want to consider is after a lot has already been discovered and some societal collapse occurs, such that people have the knowledge but not the means to test it improve upon it or discredit it.

Like the people in the middle ages who must have looked at the towering Colosseum made out of concrete, and wondered how since they had lost the recipe for concrete and the architects to use it.

A similar thing may happen in the future where there are books written in libraries where the people blindly follow instructions not knowing why.

  • 18
    $\begingroup$ As brilliant as Aristotle was, it's exceedingly myopic to say he knew everything about everything...he knew alot, but it was related to what Greeks knew in their culture. The world was far more than Greece in 300BC. I don't mean to nitpick, but the claim that anyone person knew everything ever is silly. Outside of that I entirely agree with your post $\endgroup$ – Twelfth Dec 10 '14 at 21:22
  • 8
    $\begingroup$ I thought Leibniz was the latest polymath. $\endgroup$ – celtschk Dec 10 '14 at 21:50
  • 30
    $\begingroup$ @Twelfth By "Knows everything about everything" I did not mean it literally, I meant that there was (almost) nothing left for him to learn from his peers that he had access to. There was no university course that he could take that he was not qualified to teach. $\endgroup$ – user288447 Dec 11 '14 at 0:15
  • 1
    $\begingroup$ Question, as I understand it, is not to have a single person knowing all knowledge known to humanity (we are way past that), but sum of app people knowing sum of all knowledge of humanity. @David Mulder has reasonably approximated answer. This one is not, IMHO, even if highly rated for some reason. $\endgroup$ – Peter M. Dec 11 '14 at 23:14
  • 1
    $\begingroup$ Aristotle may have "known everything about everything," but nearly everything he "knew" turned out to be wrong. But because he "knew" so much and was the ancient world's foremost authority on everything, new science that disagreed with Aristotle tended to be rejected on that basis alone. When you look at the history of medieval science and the beginning of the Scientific Method, it's a bit horrifying to see how many times this happens while building up the basics of many modern scientific fields! $\endgroup$ – Mason Wheeler May 1 '15 at 20:42

While a fascinating idea, the problem itself will never arise because of specialization.

XKCD: Purity | On the other hand, physicists like to say physics is to math as sex is to masturbation.

As soon as we know "too much" in a field, there will be new institutes on universities faculties who specialize in fundamental research. Every studying technician/engineer (for example) already has to learn "basics" like Math, Physics, Chemistry and Mechanics. If you are going to become for e.g. a higher building engineer, you will learn statics to calculate the needed dimensions for building parts like columns, etc., but you won't get into the depth of mechanics as statics is just a high level implementation of mechanics. You need to know the basics of the fundamental parts, but you don't have to know mechanics inside out to being able to calculate building statics.

When you ever thought about studying some field like Physics, you will find out that you are able to study Physics on different universities (like technical university, humanistic, etc.). This is because there are already different main topics for such general fields - example: Astro Physics, Technical Physics, General Physics, Atom Physics, etc. And that means that there will always be people who do fundamental research, specialized research and highly specialized research where one work builds upon another.


To answer the following part of the comment:

AstroPhysicists still need to learn Newtonian physics and electromagnetism. Wouldn't that specialization continue, causing say... Brown Dwarf AstroPhysicists to learn some fraction of general AstroPhysics and then their Brown Dwarf specialty? We learn tons about Brown Dwarves, and then need more specialities, with more foundational learning?

As we are the internet, we first have to fix the assumption: There won't be any specialization like "Brown Dwarf star". There might be a specialization for stars in general. And there might be people dedicating their research to only dwarf stars and maybe even only brown dwarf stars, but not as a research field that is run on an institute of an universities faculty like astrophysics. Every bigger institute has one or two leading professors, some professors who teach parts of that field and then some phd students as assistant professors. And everyone of those persons will have their research field where they do research and publish papers.

Now the point is that we only need the papers. Because rule:

You don't have to know everything. You have to know where to find something.

And this leads us to the problem that arises at the end (that we already have):

How do we organize data in a searchable fashion that brings up everything that is relevant to the question ?

The answer to that is search engines and their internal algorithms. And systems like this network: StackExchange - which has the following general idea:

Imagine StackExchange as a library where you don't get inside to search for books, but where you stand at the door and shout a question, then get handed a stack of books sorted by relevancy to your question - peer reviewed.

  • $\begingroup$ Sure, but such specialization can only go on for so long, right? AstroPhysicists still need to learn Newtonian physics and electromagnetism. Wouldn't that specialization continue, causing say... Brown Dwarf AstroPhysicists to learn some fraction of general AstroPhysics and then their Brown Dwarf specialty? We learn tons about Brown Dwarves, and then need more specialities, with more foundational learning? $\endgroup$ – Telastyn Dec 10 '14 at 16:08
  • $\begingroup$ @Telastyn As comment does not offer enough chars to answer your comment, please see the edit. $\endgroup$ – kaiser Dec 10 '14 at 16:33
  • 2
    $\begingroup$ I can't personally agree with the edits. Data is not knowledge, and even knowledge is not understanding. And even if we could quickly gain understanding of topics, it won't matter if we keep making new (dependent) topics. $\endgroup$ – Telastyn Dec 10 '14 at 16:39
  • 1
    $\begingroup$ @Telastyn Data is the base upon we do research and gain new knowledge. We assume that a base is correct and works: But we only know something until we know something better. That is the main difference between knowing and believing. And that is the reason why we quote and link sources that we based our research. We offer readers the possibility to find thing that we assumed wrong or that are already outdated. If we need to drop a topic because we found out that the base was wrong, then well... we will drop it and start from scratch. That's how science works. $\endgroup$ – kaiser Dec 10 '14 at 17:07
  • 2
    $\begingroup$ @Telastyn: One very good example of specialization is computer engineering. While it itself is a very specialized field, it has been stated that no single person on earth completely understands how computers work in order to create everything from scratch. The specialization is such that people have entire careers just adjusting recipes for doping silicon to make various semiconductor parts. I have personally designed my own CPU and created my own programming language. Built websites and program devices for monitoring the power grid. But even with my experience I dont/cannot know everything. $\endgroup$ – slebetman Dec 11 '14 at 8:24

I think one of the biggest things that help us is the concept of the 'black box'. We use it in programming all the time.

We have something and we know what to put into it, and it gives us an answer back. We don't know what goes on inside but it has been tested by the original writers. As a programmer over the last 15 years the tools keep improving and I need to write less code to get bigger, more impressive and better programs out.

Take Graphing calculators, someone has programmed them to be able to do calculus and as such you don't need a deep understanding of calc to be able to use it, just a decent general understanding of how to use the formulas correctly. That does not mean that no one needs to bother learning calc, because it is still useful and new discoveries are still happening in Math.

You see examples of this all over the place, you don't need to know how GPS triangulates satellites (or even how satellites work) to be able to create an app for a phone that does something with locations and a map.

There will still be people who will continue to work on the 'black boxes' but you don't need to know how they work in order to use them for new and interesting applications. Actually often once they have been abstracted out it makes it easier to do so. Some times the new ideas need to be sent back to have the black box redesigned to add or modify it.

In software engineering, most software is developed using high level languages, but there are still people using assembler to do different tasks. Designing better CPU's still needs an incredible understanding of machine level languages and logic gates, though I expect they have their own set of 'higher design tools'.

So in conclusion, I don't think that new discoveries will stop because we take too long learning all previous knowledge that got us there.

ETA Adding a point, for how this could happen where we can't learn enough to make new discoveries. If humanity falls into decadence and gives up trying to find new discoveries.

  • 2
    $\begingroup$ @Telastyn Exactly. but at one time you'd need to know all of that just to write a simple program. $\endgroup$ – bowlturner Dec 10 '14 at 16:01
  • 5
    $\begingroup$ @Telastyn Your question asks if we would ever spend to much time learning to make advances. My post was to show you don't need to learn everything that goes before in depth in order to 'stand on the shoulders of giants', for that is the idea of standing on their shoulders means. $\endgroup$ – bowlturner Dec 10 '14 at 16:08
  • 2
    $\begingroup$ But those abstractions aren't 100% free. Even if you're not learning them in depth, there's still time spent learning them. Unless lifespans outrace that added time, the learning will win. $\endgroup$ – Telastyn Dec 10 '14 at 16:12
  • 6
    $\begingroup$ @Telastyn only if the abstraction process isn't able to keep up. My bet is it will. $\endgroup$ – bowlturner Dec 10 '14 at 16:26
  • 2
    $\begingroup$ @Telastyn: While the compiler writer needs to have an understanding compiler technology, he in return doesn't need to understand cryptographic algorithms. And when compilers get too complex for a single person to understand them completely (I'm not sure if we haven't already reached that level), then more specializations will occur: People who are specialized on a certain front end (e.g. there will be C++ front end specialists, or Fortran front end specialists), others who are specialized on a certain class of optimizations, and again others who are specialized in specific backends. $\endgroup$ – celtschk Dec 10 '14 at 22:09

The implication here is that each individual discoverer must start from nothing but a bag of crying cells, and build up knowledge in a linear order before making a discovery in a vacuum.

In reality, I find we have an entire interwoven society trying to make the discoveries, not independent individuals. There is an entire section of society dedicated to distilling the human essence into teaching. There is an entire section devoted to building infrastructure to make it easier to step beyond. There is an entire section devoted to getting discoverers together, so that they don't ALL have to learn ALL of the knowledge; they merely need to have all of the knowledge when they put their minds together.

Consider that the trade knowledge needed to run a particle accelerator is equally essential to discovery as the quantum physics models used to point the accelerator in new and exciting directions. The physicists probably don't know how to correctly shim the hundreds of segments of the accelerator to be in a perfect shape (and doesn't have the time to learn). The physics probably hasn't spent enough time with high voltage to wire up thousands of electromagnets without a short taking the entire accelerator down. This knowledge, held in the minds of the tradesmen who support the physicists, is equally essential but the physicists never had to learn them; these skills were learned in parallel by all of humanity.

The only thing I have found which can leave us with no time to discover is society itself. If society dulls, and our lives suddenly require an entire lifetime of learning just to survive, that could be the cusp where humanity simply cannot learn any further. However, even then there is a light at the end of the tunnel. The poets have a long list of skills like "how to love" which take a lifetime to learn, and yet we keep working on them day after day. Perhaps one day, discovery will simply take the form of loving the universe and seeing what it wishes to tell us today.

Oh fine! Lets see some math

Lets try to put some mathematical equations down to make sure we're all on the same page. I'll use it to show how a rather boring society resembling the Vulcans could go about never ending discover

First off, I am going to assume there is a never ending supply of things to discover in the universe. If there is a finite number of things to discover, then it is trivial to show that the number of discoveries humankind can make is finite. Let us define the universe of potential discoveries to be $\mathbb{D}$

I am going to assume the only thing in our brain that matters in the long run are structures. These are structures you have to learn over time in order to effectively do a task, such as discovering a new direction. I believe there is more to the brain, but I think this is close enough to model your question of learning and technology. Let us define these structures to be $\mathbb{S}$, the set of all helpful structures that the human brain can possibly organize into, and let $\text{Fits}(S), S\in \mathbb{S}$ to be a predicate that returns true if the set of structures $S$ would fit into a single human brain, and false otherwise. Because entering the world with new structures makes it trivial to prove we can keep discovering, we can assume $S$ of a newborn is $\emptyset$.

Now we need a notation for learning. I will assume, for simplicity, that people learn at a constant rate through their entire lives. I leave it to the reader to show that handling the case where learning rate is variable is a trivial transform from this simpler case. Because I am arguing that we will never run out of things to learn, I can assume the worst case of "you can only learn one thing at a time" without loss of generality. Consider the universe of learning activities, $\mathbb{L}$. For any learning activity $l \in \mathbb{L}$, we can define a function $\text{cost}_{\text{learn}}(l, S)$ which defines the cost (in time) of doing learning activity $l$ given that you already have all of the structures $S$ in your head. Let $\text{results}_{\text{learn}}(l, S)$ be a function which returns a set of structures in your brain after doing a learning activity.

Finally, we need a notation for discovery. $\text{cost}_{\text{discover}}(d, S)$ is the cost of discovering a particular element of $\mathbb{D}$.

Now we can define the goals. Let us define $\text{cost}_{\text{schooling}}(L)$ and $\text{results}_{\text{schooling}}(L)$ where $L$ is an ordered set of learning activities to be the cost and results of raising an individual up from $S = \emptyset$ through a sequence of learning activities. Thus $\text{cost}_{\text{schooling}}$ will be the sum of $\text{cost}_{\text{learn}}$, and $\text{results}_{\text{schooling}}$ will be the final result at the end of iterating $\text{results}_{\text{learn}}$. Our goal is to prove that there can always be a $\text{cost}_{\text{schooling}}(L) + \text{cost}_{\text{discover}}(d, \text{results}_{\text{schooling}}(L)) < \text{lifespan}$. Let us assign this a predicate: $\text{DiscoveryCapable}(L, D_{prev} \Leftrightarrow \exists_{d\in\mathbb{D},L^\prime}[(\forall{l\in L^\prime} l\in L)\land d\notin D_{prev}]$, which is a mouthful to day "A society is DiscoveryCapable if, for their set of known learning activities, and previously discovered disoveries, there exists a discoverable thing." Let us also add $\text{Discoverable}(L, d) \Leftrightarrow \exists_{L^\prime} \text{cost}_{\text{schooling}}(L^\prime) + \text{cost}_{\text{discover}}(d, \text{results}_{L^\prime}) < \text{lifespan}$, or "A discovery is discoverable if, given the known set of learning activities, someone can discover it in a lifetime."

Now here we will note that $\forall_{l\in\mathbb{L}}l \in \mathbb{D}$, or in words, every learning activity is something which can be discovered. This leads to a "Lotus Eaters" situation, where could simply continuously develop new ways to learn without going anywhere, so lets fix that. Lets define $\text{Trivial}(l)$ to be true if $\forall_{S\in populace}\exists_{L_0} (\forall_s s\in \text{results}_{\text{learn}}(l, S) \to \text{results}_{\text{learn}}(l, S_0)) \land \text{cost}_{\text{learn}}(L) \ge \text{cost}_{\text{schooling}}(L_0) $. In other words, its trivial to develop a new learning activity which doesn't teach anything new and costs more than an existing schooling!

Now we do a proof by contradiction. We assume $\text{DiscoveryCapable}(L, D_{prev})$ is false for our society. We will prove this is contradictory, meaning there is no such society that cannot find a discovery.

If $\text{DiscoveryCapable}$ is false, then that means there are no new non-trivial learning activities which are discoverable. If we find that there must be a non-trivial learning activity to discover, we have a proof by contradiction. This means we must prove $\forall_{L, D_{prev}}\exists_l \lnot \text{Trivial}(l) \land \text{Discoverable}(L, l)$

Consider the Turing machine, which is accepted to be far simpler than even a human. If we can prove that, at this time, a Turing machine can develop a new useful learning activity for us, then we can make a discovery simply by following that program. We are, after all, at least as impressive as computers.

Let us devise a turing machine to help. Select a subset of $L$ called $L_T$ which is the learning activities which can be analyzed by a Turing machine. We want to find a program which finds a $l \notin L_T$ such that $\lnot \text{Trivial}(l)$. The first step is easy. It is trivial for computers to find an activity $\exists_{l\in 2^{L_T}} l \notin L$. Such power set behaviors occur all the time in NP problems.

Now what if the computer can't do this? The next step is to gather some data about the universe. If we can't find any new data, then we are literally out of things to discover. If we find new data, we can have the computers crunch it harder, to find things that we don't understand, but computers can find. If they cannot, then all Turing-capable learning methods are exhausted, and we have covered the universe with our computational prowess. We, in effect, used computers to extend our life, crunching a subset of our possible learning activities, in hopes of finding a new one.

And now we sit back and look at the non Turing learning activities. It is not easy to tell if there is a faster way to learn such things. In fact, the only limit seems to be creativity.

The only limit for our capacity to discover is our own creativity.

  • 1
    $\begingroup$ Certainly a single individual need not know how to build the particle accelerator, but the physicist who proposes the Higgs-Boson needs to have solid mastery of the underlying particle physics (and its pre-requisites) to theorize its plausible existence. $\endgroup$ – Telastyn Dec 10 '14 at 15:56
  • $\begingroup$ Is there any reason he could not rely on the help of his fellow physicists? Consider that the PRL Symmetry Breaking Papers were authored by not just Dr. Peter Higgs, but 6 physicists. $\endgroup$ – Cort Ammon Dec 10 '14 at 16:06
  • 2
    $\begingroup$ @Telastyn: If you accept the ideas of science fiction as potential sources for an answer, consider the Drummers from Neil Stephenson's "Diamond Age." They are an example of a distributed mind which could theoretically discover without bounds. $\endgroup$ – Cort Ammon Dec 10 '14 at 16:08
  • 2
    $\begingroup$ Alternatively: consider the field of computer science, where 99% of programmers cannot program in assembly, and know little about cache coherency, but continue to discover new ways to put code together. $\endgroup$ – Cort Ammon Dec 10 '14 at 16:09
  • 1
    $\begingroup$ @Telastyn: Of course to come up with the Higgs boson you need to be familiar with particle physics. You also need to know quantum theory, but already here you can make tradeoffs: I'm pretty sure Peter Higgs didn't need detailed knowledge for example of the measurement problem, or of EPR, in order to come up with the Higgs mechanism. And of course you need natural, real and complex numbers for doing quantum mechanics, but that doesn't mean you need to know how to define the real numbers using Dedekind cuts or nested intervals, or the definition of natural numbers from ZFC set theory. $\endgroup$ – celtschk Dec 10 '14 at 22:25

Sometimes the most significant contributions to science are ways to simplify or visualize things to make a complex topic easier to learn and understand. Other times the contribution is a way to organize knowledge again making the body of scientific knowledge easier to learn and understand.

So far in human history ways have been invented to help learning so that the best and brightest are able to contribute new discoveries after learning the base knowledge of their field of research. I see no reason this trend can not continue.

  • 1
    $\begingroup$ +1 for a perfect example - I just knew "simplify or visualize" linked to Feynman diagrams before I even moused-over the link. $\endgroup$ – DeveloperInDevelopment Dec 13 '14 at 17:10

The premise is flawed.

Your premise assumes knowledge is constantly increasing. Actually, evolutionary scientists assure us that mankind 5000 years ago had precisely the same intellectual capacity as mankind today. The total amount of knowledge that any one person has doesn't change at all.

You know nothing that is more important than what a man of 5000 years ago knew - in a pure or abstract sense. You know things that pertain to survival in your world, he knew things that pertained to survival in his world.

Your example was Calculus. Calculus is not useful in almost any world, its useful only these days to people who program calculus engines in computers, since anyone who needs applied Calculus would use a computer, and doesn't need to actually understand anything about Calculus.

And a fellow throwing a spear 5000 years ago didn't need Calculus to predict its trajectory. As an example, I just did a complex image processing problem involving which required, at the computational level, some really advanced matrix algebra, but I know nothing about such things and don't need to know them. I used a "perspective transformation" that someone else had written. I understand at a rudimentary level what it means to transform a perspective in a picture, and then I used the routine someone had written.

The only person on earth who needs to know matrix algebra is the guy writing these tools. I just used it, at the cutting edge of vision processing, at the cutting edge of human achievement, while knowing nothing about that ancient and pointless science or math underlying it. My knowledge is exactly the same as the guy 5000 year ago. He knew all about hunting, how to get downwind of his quarry, how to find his quarry, how to kill it.

His knowledge of such things was advanced. Its kind of arrogant and feckless to imagine people today know more than people in the past. Its a very common kind of cultural arrogance, to imagine one's tiny bit of information is the best bits of information that exist. Its the proximity fallacy, because you know those things, you imagine they are the only things that are true.

Have ore respect for your ancestors. They knew a lot more about all kinds of things than you know. Such as philosophy, religion, the meaning of life. They were far more advanced in these areas.

  • 1
    $\begingroup$ But what if you pluck a mature, experienced hunter gatherer from some isolated tribe and give them a modern, Western education? Or take a physics postdoc and parachute them into the Amazon rainforest? Assuming we aren't being racist or ageist about this, each of them could learn the other's skillset in addition to their own. Surely this is "more" knowledge? There are quite a number of survivalists, military personnel and astronauts with similar skillsets. It is not right to be dismissive of ancient knowledge, but we can also show it too much reverence. It is not beyond us. $\endgroup$ – DeveloperInDevelopment Dec 11 '14 at 19:37
  • 2
    $\begingroup$ Nice answer, could use some paragraphs though.... $\endgroup$ – Tim B Dec 12 '14 at 12:57
  • $\begingroup$ You don't have to suppose that people today are smarter than people were in the past to say that there is more total knowledge. There's lots of information available today that was not available 1000 years ago. Some information that people knew in the past has been lost, but not all that much. Sure, I don't have any idea how to successfully hunt and kill an animal. I've never hunted anything in my life. But there are plenty of people in the world who do know how. If I wanted to know, I could find out. But there was no one 1000 years ago who knew how to build an electrical generator. $\endgroup$ – Jay Jun 22 '15 at 5:29

Specialization as the 'solution'

So, most answers are bringing up specialization as the 'solution' to the 'problem', however this is ignoring that despite specialization there is a limit. After all, right now it takes about 20 to 30 years to get specialized in a cutting edge specific field where actual research is done. If that would be the average age of humans than we would thus have reached the point already. Luckily we get about 4 times as old (though not all of that is actually 'useful' time), so there is still a lot of room for even more specialization and our primary and secondary school systems are hardly specialized as well, so there is still room for improvement there as well.

Now, we can go out on a limp and imagine things like AI's taking over the work for us or increasing our longevity, but that presupposes that such developments could be made before we hit the boundary under investigation. Whether or not that's likely is hard to determine, but from a world building perspective both things are quite possible and arguing one way or the other is quite pointless.

So, can we build a world that hit such a boundary?

If we would like to argue that this boundary was reached and estimate at which point this would be the argument would look something like the following:

  1. 'Knowledge' is growing quickly, making any estimate is hard, but if I take my own fields an estimate that knowledge is doubled over a time period of the last 30 years would seem quite conservative.
  2. Right now it takes 20 years conservatively to get to the cutting edge of a certain field
    • Note: I am counting child years as well, why? Because that's efficient learning just as much as later stages in life are
  3. At this rate it would mean that in another 30 years without any increases in efficiency it would take 40 years to get to the cutting edge and in 60 years it would take 80 years.
  4. Education will improve in efficiencly a lot the closer we reach this boundary, so for a conservative estimate 150 years would sound quite reasonable
    • (Of course still assuming no gigantic breakthrough is made which removes this issue entirely, but within the context of worldbuilding.SE that's quite possible).

I believe this will never happen. After all, that's why we have specialists. For example, I study approximatly 5 years IT, then call myself an expert in IT (compared to the average citizen). I have no idea about Physics (except what I need), Biology etc.

Ask a Microbiologist about ... something not being Microbiology. He will most likely only be able to give you a vague answer. It will still be better than what I ever could give you, but you would be better off asking the right specialist. If I would get a dollar for each time someone comes to me and asks "You study IT, right?"...

So no, I don't think we will reach a point where we can't advance because we know too much.

But, we might reach a point where an advancment in one area doesn't reach another area, simply because they became to distant. Each area of Science speaks its own language. When I talk with IT-Students from other universities, I often have to clear some words which are in itself Synonyms, but one is common at one place and one at the other. Now imagine a wide field as Physics. Two fields of research could differ so much that both use the same words with different meanings, and different words with the same meaning. It would be impossible for a researcher A to find material from area B, simply because A doesn't know the right key-words.

Luckily we have computers today. Did you ever google something, and got results which did not contain any of the words you searched, but were still on-topic?
Computers are fast in comparing huge amounts of data. Google for example rates content depending on it's importance (how many people use this specific ressource?) and its relevance (how many of the search words or its synonyms fit this result?).

  • $\begingroup$ Ah, you brought up specialization as a reason why it wouldn't happen, which is ignoring the problem entirely. Despite specialization the question still applies~ specialization just pushes the point back a lot, but it's not an answer at all, because without specialization the question wouldn't have rised up in the first place. $\endgroup$ – David Mulder Dec 12 '14 at 12:53
  • $\begingroup$ @David In my opinion this is not totally right. We are more and more outsourcing our problems. Search Engines, Data Bases, Automatic Solvers ect takes a huge load of problems from our shoulders. In the future, progress in a field may slow until this infrastructure is in place, but then will continue again. I doubt we will really reach a point where installing this infrastructure becomes impossible. If you want to discuss this further, let us move to a chatroom. $\endgroup$ – J_F_B_M Dec 12 '14 at 14:16
  • $\begingroup$ Nah, you were asking why the downvote, so wanted to explain. Personally I think the existence of the limit is a given, the question being how far of it is. $\endgroup$ – David Mulder Dec 12 '14 at 15:23

Going to use the concept of Holon's by Kevin Wilbur for this one...the fact is all knowledge is somewhere based on assumption and no one by definition can know all.

Every idea is a 'holon'. It is in and of itself a whole (one whole idea). However if you zoom in on this holon, you will discover it is not a single whole, but a whole made up of other parts. Each of these parts is a holon in and of itself. So all ideas are holons and are composed of holons. Now if you 'zoom out' you will discover this holon is indeed part of other holons as well. Kinda a crash course on the Holon concept. All knowledge is based on one of two methods...dissecting existing holons to discover new ones or combining holons to make new ones.

What this ultimately means is all knowledge (always) is somewhere based on assumption (whether we convince ourselves this assumption is well established by the scientific method or otherwise) where we are using a holon without understanding the elements that fully make it up. You don't need to understand every level of a holon to create a new one (otherwise knowledge by definition is impossible). A self driving car can be created by someone not knowing the basics of internal combustion, they just need the concept of the car. Combustion engines can be created by those that do not understand the chemistry behind combustion, they just need the concept of combustion. One who understands the chemistry behind combustion doesn't need to understand the mechanic of a combustion engine, nor the electronics behind a self driving car, nor the physics of the atoms structure involved in chemistry. Do you need to understand the inner workings of the monitor in front of you to be able to use it to create new ideas?

So my answer is no...the very foundation of knowledge is ultimately assumption, as the human race advances, the starting point of assumption for each human changes. One can argue that a holon can exist that know one person could ever fully conceive, but our social nature allows us to adopt another's holons (and assumptions) as required.


There are several answers here that seem to deal with specialization and to some other extent the idea that most discoveries are teams of scientists now-a-days.

However I would like to point out the fact that as knowledge increases (where you have to "learn more" to be at the top of your field) so too does the general quality of life for humanity. As we develop better medical techniques and practices life expectancy will continue to improve.

Some of the more science fiction themes on this topic involve ideas like clinical immortality or the ability to transfer our biological consciousness to a machine; these are a ways off in terms of what science may be capable of currently but in the future may be common place.

For a quick example of this concept look at the setting in which the TV show Futurama takes place: The Professor is over 159 years old (at the beginning of the show) and spent a great deal of time learning about science, however he's still alive and inventing things even at that age. We also see the technology to keep heads alive in jars which can limit the ability to work on some things (although we do see the capability for the heads to interface with robot bodies) and several episodes show the heads of musicians still performing and writing new works.


I'll try to take a different tack, since your comments suggest no one's really gone where you'd like yet. There's similar concern which basically can be summed as, "given the rate machines are overtaking work humans have traditionally done better, what will we do when there aren't enough necessary jobs left to keep us busy?"

It's a little surprising to me that these parallel as well as they do, since I suspect the question I pose may become a reality in our lifetime, while I found myself agreeing with most of the respondents to your question that there's still so much we don't know about ourselves and this planet and the universe that we'll have much left to learn.

To illustrate this parallel, I'll reduce both statements to: Given that our drive to improve is one of our more endearing traits, how would humans cope with a situation in which there isn't really anything meaningful left to accomplish?

Amusingly, I think generalizing both of our concerns into this form reveals the only aspirational escape from both: If our species survives until the educational investment required to advance our traditional fields is longer than the time it would take to fly to the nearest unexplored star/planet, manifest destiny will once again carry the most curious of us off into the darkness on its wings.


In Science Fiction, civilisations where nothing new is created can be found - for example, Asimov's Trantor civilisation.

However, in that case it still seems that the problem is not that it takes too long to learn everything before you can start something new, but that doing your own, new thing instead of just regurgitating many hundred year old texts by the great masters of the past is frowned upon.

If we reached the point, I think there might be a huge phase of "cleaning up" to do, where all the knowledge that we have collected and that someone might want to learn is re-examined, checked for correctness, and put into the form that is easiest and quickest to learn. At the moment, scientists don't have the time to put their works into an easily accessible form as they need to publish; so if it takes too long to go through the whole works that need to be known to find something new, there's plenty of work to do to reduce the time of learning.


Forgetting something is also a part of solution.

Example: One of the almost-forgotten, recently revived arts is Polynesian navigation, without charts or navigation equipment. It is nice that it was saved and someone is trained in it, but if forgotten, it would have no effects on our capability to navigate.

So forgetting less effective way to goal, if we have more effective way, buys you some wiggle room.

Possibly, in the future we develop more advanced/specialized ways to accomplish same design goals. Like integrating some measurement instruments directly into our senses, so you can learn selected areas faster. Of course this would push specialization even deeper. Better mental models of some laws may also improve speed of learning.

This improvements, however marginal, will push the break-even moment further - eventually we may integrate cyber chips to brain and became cyborgs. Sci-fi area. Singularity ensues.


Interesting question, but not really askable.

Knowledge is an efficiency modifier or force multiplier for discovery, not a necessary element. It will certainly help recognize discovery, but the discovery process is often counterintuitive to those steeped in knowledge.

And the assumption that people in a society will absorb the knowledge on which it is based... Not really viable. If it were, would we have Creationists?

  • $\begingroup$ For some things maybe. But if I have no idea that computers (or some analog) exist, I'm not going to discover hash tables. It might be counterintuitive to people with knowledge, but it is literally unthinkable without enough pre-requisite knowledge. $\endgroup$ – Telastyn Dec 12 '14 at 17:41
  • $\begingroup$ I think that (having creationists) is a matter of the educational system and presense of viral memes. Addressing that would be progress but still not unlimited progress, and it doesn't really matter to one who is learning. $\endgroup$ – JDługosz Dec 13 '14 at 2:58

There is also the possibility of a more pure knowledge to be gained about the stuff we thought we knew pretty well. As we go, subjects may become easier to understand. Just like how Algebra used to be incredibly hard and mostly in the dark, and only the most brilliant worked on it, but now that we know more we are teaching our middle schoolers it.

This stuff is theoretical, and I like to refer to it as the Compaction of Knowledge. It's kind-of like describing a game by starting out with its source code (and you don‘t know how to read it). Maybe we will be able to describe the entire universe in a set of words some day, when we understand why things are the way they are.


We should think of this problem in terms of trees. For simplicity's sake, let us assume that knowledge forms a directed acyclic graph from the most basic to the most complicated facts. Further, let us consider that "learning" is a means of gradually covering this graph, starting from the root. A "discovery" occurs when someone reaches a boundary of the graph, and extends it with a new node.

Part of what makes discovery difficult is that it often requires covering large parts of the graph to correctly extend it, and we often don't know which parts to study in order to make an advance (not to mention that luck/serendipity surely plays a role). Even so, the essential feature of this tree-like graph, with respect to the posed question, is: how quickly does it branch? We can roughly consider the depth of the tree at any given node to be a rough measure of how much study is required to comprehend that node.

If the tree branches only narrowly, then the depth of knowledge required to add a new layer grows quickly, and we should anticipate the limits of human discovery. If, on the other hand, it branches widely, then the increase in learning required will grow much more slowly than the number of discoverable facts available at a given knowledge level.

Since tree depth (or height, if you prefer) only grows logarithmically with the total size of the tree (for regular/full trees), we have some comfort that the universe of discoverable facts grows much more quickly than the need to study to reach those facts. On the other hand, it implies that we have an exponentially rising number of discoverable facts to learn. ;)

As some people pointed out, the tree isn't exactly "clean" or "neat". At some point, the bigger danger will likely be folks rediscovering things because the similarity or equivalence of some concepts will not be readily apparent, or because the knowledge required to understand them will not be available to the independent discoverers.


It is called singularity.

There is theory (I forgot name or link) of increasing rate of progress.

Before and during stone age, adding one step to technology took 10K years. After few steps, we got to next level, and rate of progress increased by two magnitudes about 12K years ago, to a century or so.

After about a dozen steps, say 1300CE, rate of progress increased again, to under a century. Few steps, industrial revolution, next level, rate of progress increased by another order of magnitude: step is about one generation. At it was about one generation since, because it takes 20+ years to start learning new stuff.

Most recently, as you correctly noticed, rate of progress is so steep that often is impossible to even read all relevant literature in your area, and some research is duplicated because of that.

In few decades, when artificial intelligence will become capable to design/improve AI, rate of progress will increase again. AI does not need to spend 10 years to get born, potty trained, learn to talk and become literate. And humans will become incapable of compete, or catch up. AI will be able to do research faster than we can read or comprehend it.

Interesting times ahead. But we do have few decades to prepare for that.

Only protection is not to let this happen, I do hope that we will protect somehow ourselves from such AI. But AI might be smart enough not to let us know what it can do. And having smart AI out-innovate our enemies might seem to be a military advantage, for a while.

  • 1
    $\begingroup$ Actually, that's not what the singularity is usually described as. The Singularity is when tech progress goes infinite in speed, not when it stops. $\endgroup$ – Oldcat Dec 10 '14 at 18:47
  • $\begingroup$ Progress will go faster for AI than humans can understand and learn (even if not infinite). It will not stop, and I never said it will stop. Why do you think I said progress will stop? $\endgroup$ – Peter M. Dec 10 '14 at 19:46
  • 1
    $\begingroup$ More I learn about computers (and its a lot) the less I think AI will take over the world. Heat limits. Power limits. (Same as Heat really). Limited latency and bandwidth of communication to other nearby compute nodes. Physics limits on feature fabrication. Money limits on building new fabs. Combine the whole with hardware failure rates. (There are already supercomputer clusters limited in size only because the components fail faster than the staff can replace them.) $\endgroup$ – Zan Lynx Dec 10 '14 at 19:55
  • $\begingroup$ That's why we need AI to design better components :-) $\endgroup$ – Peter M. Dec 10 '14 at 19:59
  • $\begingroup$ What you describe isn't the same thing that others call a singularity. Singularity is the absurd/wrong/fantasy concept that all human knowledge and experience can and will be replaced by a huge computer system that is the sum of all those things and more, but that only makes sense if you falsely believe that all humans are are data and logic, AND that any machine could meaningfully represent and replace and make sense of all of that, etc. What you propose is more like the "computers will get better than us yet still be conquerors and replace us" horror idea, which is slightly more plausible. $\endgroup$ – Dronz Dec 10 '14 at 22:16

There are many excellent answers for why this probably wouldn't happen.. for an incredible amount of time. One thing I'd like to add to that discussion is that our life expectancy will continue to grow with more knowledge of the medical field. It could be eventually that we can spend 200 years learning a topic, but we don't die until we are 1000 years old.

But perhaps, in the extremely far future..
after there is nothing left to discover about our physical universe
after all physical items are perfectly designed..

It could be that the only thing left to "discover" is who we are and to experience the things that we, as individuals, haven't yet experienced. The only things left to create are what is considered to be art today, things that create new unique experiences.


Something being missed here is that this can happen to a specific field.

No one will ever know everything.

But there can be fields where no further progress is possible because it takes a lifetime to develop an understanding to the point where you can actually add to the corpus of information.

I remember reading particle physics is about there.


Sure, I can come up with 4 arguments for how this would happen:

Premise 1: It takes a finite amount of time to read something.

Premise 2: It takes a finite amount of time to write something.

Argument 1: You need to learn a set amount of prerequisites before you can discover new things. As the amount of prerequisites grows, eventually merely reading the prerequisites will take longer than the human lifespan.

Argument 2: You need time to write down your ideas and discoveries. As words are continuously being written, the range of the English alphabet is continuously used up, eventually merely writing something that hasn't been written before will take longer than the human lifespan.

Argument 3: You need new equipment to make new scientific discoveries. As the space of possible equipments are continuously exhausted, eventually it will take longer than the human lifespan just to invent a new equipment.

Argument 4: Ideas can be represented as a sequence of bits. As the space of ideas is continuously being used up, eventually it will take more bits than the number of atoms in the universe to represent a new idea.

  • 1
    $\begingroup$ If my math isn't massively wrong it would take over 11.41 million years for 10 billion humans typing at 100 words per minute for 16 hours a day, 365 days a year, to exhaust all 4-word permutations of the estimated ~1m words in English, assuming all the neologists can keep their mouths shut. The same source estimates we neologize every ~98 minutes. The first neologism alone would add 45.6 years to our sentence, and it would only get worse from there. $\endgroup$ – abathur Dec 11 '14 at 4:15

protected by HDE 226868 Dec 13 '14 at 1:25

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.