From 1906, as quoted by Albert Bigelow Paine:
On the trip down in the dining-car there was a discussion concerning the copyrighting of ideas, which finally resolved itself into the possibility of originating a new one. Clemens said:
There is no such thing as a new idea. It is impossible. We simply take a lot of old ideas and put them into a sort of mental kaleidoscope. We give them a turn and they make new and curious combinations. We keep on turning and making new combinations indefinitely; but they are the same old pieces of colored glass that have been in use through all the ages.
See here for further discussion of that topic (thinking of much of our own speech and writing as a form of sequential rearrangement of existing material). I do think that this sort of mechanistic analogy can give us insights into our own processes of speaking and writing.
Also relevant is this article, Artificial Intelligence and Aesthetic Judgment, with Jessica Hullman and Ari Holtzman.
As a first reaction, I don’t agree with Clemens’ statement (at least, not from the limited context provided). We definitely build upon ideas we get from elsewhere; but I think our interpretation of these ideas is always going to be partially unique because of the experiences we’ve had to that point. These experiences are a function of our biological makeup and our environment, and though that particular combination may be similar across many individuals, it will never be identical. In short, I think it might be better to characterize our speech and writing (and thinking!) as a noisy form of sequential rearrangement of existing material; and from that noise can come new insights and ideas.
Max:
I read a biography of Twain many years ago, and one thing I don’t remember is whether everyone called him Sam, or did people who knew him later in life call him Mark? That is, was it just a pen name or did it become the name that he went by. I have a similar question with Orwell. Did any of his friends call him George? His second wife took the name Orwell so that suggests that it became his name; on the other hand, his son’s last name is Blair. Cicely Fairfield took the pen name Rebecca West, but then her son was Anthony West, so I guess that she took the name West in her life as well as her writing?
Perhaps they rearranged the names into a sort of mental kaleidoscope and ended up calling him “Sark”. (Or “Mam”.)
“I don’t remember is whether everyone called him Sam, or did people who knew him later in life call him Mark?”
Samuel Clemens didn’t start using the pen name Mark Twain until 1863, when he was age 28. Don’t know about people who came to know him only after he started using his pen name, but here are some examples of post-1863 personal correspondence in which friends continued to call him Sam:
from William Dean Howells (author and close friend)
Date: November 15, 1880
Quote: “Dear Sam, I’ve read your piece and it’s brilliant as usual. We must get together again soon; it’s been too long.”
from Joseph Twichell (close friend and pastor):
Date: April 25, 1876
Quote: “Sam, the sermon went well yesterday, and I hope I’ll see you and Livy soon. We should take another walk like old times.”
from Henry H. Rogers (close friend and financier)
Date: December 3, 1893
Quote: “Sam, don’t worry about the financial situation just yet. We’ll find a way through. You’ve always landed on your feet.”
Also, I don’t agree that there are “no new ideas” anyway. Wasn’t Newton’s theory of gravity and laws of motion, the laws of thermodynamics, Einstein’s theory of relativity, basically every major scientific theory or discovery (as well as ultimately proven incorrect ones like phrenology or Lamarck’s view of evolution) a “new idea”? Didn’t most major philosophies and religions in history introduce at least a few new ideas, even if they also reaffirm and recombine existing moral and religious ideas? What about math – Euclid’s geometry or Newton/Leibniz’s calculus or Cantor’s infinites?
Now, *most people* of course don’t have “new ideas” on this historic scale. But even one counterexample is enough to disprove an absolute statement.
If Twain were to have said, “most people don’t have genuinely new ideas, they merely take old ideas and put them into a kind of mental kaleidoscope” etc, that would be far more defensible. But even then I wonder if it’s really true.
+1
There are two things going on. Generation and selection. My thinking is that we’re generating things very near the state-of-the-art based on what’s come before. With a lot of people doing that, some of whom are better at it than others, we can then select which things which are “right” and build on that. It’s usually much easier to verify than to generate (e.g., the generate and test formulation of the computational class NP (non-deterministic polynomial)).
Most scientific “discoveries” would be made by someone else very soon if they were not made by the original thinker. Leibniz and Newton indepentently developed a theory of calculus. Four different individuals independently introduced the optimization algorithm BFGS in 1970ish. I don’t think Einstein would’ve gotten to relativity without Riemann’s take on geometry and then all the work around the Lorentz transform in the late 1800s; see https://en.wikipedia.org/wiki/Lorentz_transformation#History.
My point isn’t that these folks weren’t great thinkers, just that it wasn’t a coincidence when and where they discovered what they did.
Ah, but then where do new ideas come from?
This can seem quite academic, but you can think about it as, what changes to my routine or environment could I make to maximize my chances of generating new ideas?
An economist of a previous generation might say that a new idea is simply a combination of old “atomistic” ideas (weights in an AI model?) that no one else had yet to stumble upon. Combinatorial explosion means the set of combinations of atomistic ideas waiting to be discovered is limitless in practical terms, and furthermore that the set of discoverable ideas explodes every time a “new” idea is discovered.
I think new ideas as a new combination of existing sub-ideas (Twain’s kaleidoscope) misses the contribution of genuinely new information. Einstein’s theory of relativity couldn’t have arisen without the previous century of observations. Darwin’s theory of evolution couldn’t have arisen without comparing life forms across different continents.
I would argue – at least by my definition of a new idea, which may not be everyone’s – exposure to new information (information seen, or recognized, by few or no people before*) would maximize the chance of genuinely new ideas.
*Or only by people with a vastly different cultural and intellectual background than yours
At least along scientific and artistic lines. no clue for math or philosophy, etc. Don’t know enough about those fields.
Said Clemens, as he ripped across the landscape in a then-modern railroad dining car, on a then-modern railroad, the existence of which required hundreds of new ideas over the previous 300 years, many having occurred during his lifetime (just a few): the 4-6 wheel railroad truck, 1834; standardized time, 1850; dining cars, 1860; bessemer process and subsequent improvements, 1850 onward. No doubt, dozens more.
Odd that he would say such a thing given his previous work on steamboats….
Chipmunk:
Good point about the steamboats! Maybe he was thinking of literary ideas.
Well…if there are no new literary ideas, if humans have long tapped all the potential concepts available to human nature then what’s the point of social sciences? To figure out what the humans already know intuitively?
I think about that alot when you and jessica talk about human perception of probability. Mostly like humans have already outwitted both of you regarding most probabilty problems. You just don’t know enough yet to know that. For example a while back you referenced a paper that claimed to discover that, frequently, humans intuitively estimate the average composition of a subgroup to be the same as the parent group, and this estimation is often mistaken. But it occurred to me later that esimating a subgroup to have components proportionally equal to its parent group is the optimal estimation to make with no experience to the contrary. If nothing else, esimtaing in the middle gives the most flexibility to respond to a mis-estimation, while an incorrect estimate on the extreme would leave people with little flexibility to shift position to meet the new reality.
Now, I don’t suppose the idea that I expressed in the previous paragraph is a “new” idea. But under the current concepts of social sciences, it’s so foriegn it might as well be a new idea. So in that sense, whether or not Clemens is right that that there’s nothing new under the Sun, the Sun is the only entity that recalls everything that has gone before, and thus the only entity for which there are no new ideas.
Er..Also I can’t help but wonder: would Orwel, Bradbury, Wells or Azimov agree that technology has not given rise to new literary ideas? I’m not aware that 1984 or Farenheight 451 draw on Shakespeare or Greek mythology, but I’m hardly versed in either. Or, if you’re familiar with the band Rush – what about 2112 (or for that matter the Truman Show, kind of a take on 1984)? Is it just Exodus revised for the Space age? The escape from tyrrany? Or is the Omnipotent State a new idea? I’m inclined to the latter. After all, until the close of the 19th Century, the earth had uncharted territory where the state had no control. The idea expressed in these works, that the state is inescapable, I believe is fundamentally new. What say you?
Hmm. I think that is a really interesting point. I don’t think the *rhetoric* of an omnipotent state is new – Mesopotamian kings called themselves King of the World 4000 years ago, but then they had no idea how large the world was! But a government with truly detailed knowledge and control of individuals’ lives is pretty modern, dependent on very recent infrastructure and technology.
So the idea of world tyranny isn’t new, but I think the 1984 style “all-seeing” State is.
confused:
Sure, yes: many have claimed to literally be a manifestation of God or have claimed the power to rule by divine right – an absolute statement of omnipotence in the eyes of their subjects. But here’s thought: past rulers claimed omnipotence loudly but in reality had very limited reach; while modern states, which are approaching actual omnipotence, would rather not talk about it so loudly and would rather their subjects aspire to percieve themselves as free.
Also I feel like “omnipotence” is not quite the word I’m looking for, nor “all-seeing”…in the modern state, the “individual” is part of a colonial entity, not just visible to the state but fully immersed within and controled by it.
Yeah, sorry to amble on, but Clemens was wrong however you consider his remarks. It’s incorrect to suggest that “literary” ideas are independent of technology. Just think of the fertile literary ground in gender transformation – inconcievable only a few decades ago and now widespread. Even species transformation is little more than a matter of legality at this point – which brings me back full circle to…The Fly. Would you say that The Fly is just a rehash of Frankenstein or Phantom of the Opera? Or are all three technologically driven forerunners of species transformation?
Chipmunk:
Just to be clear, I’m not trying to defend Twain’s statement as anything more than a witticism on his part. I just thought it was interesting as a sort of anticipation of claims about the intelligence of modern chatbots.
or you could see it as foreshadowing of the realization that humans aren’t so intelligent after all…little more than silicon chatbots mixing up word salad, having no real idea what the hell they’re talking about, just babbling stupidly in an effort to sound “intelligent” for the social benefits that go with it. IMO that’s kind of in line with Bob’s view that, if humans are in general considered to be “intelligence”, chatbots have already passed the level of “intelligence” of the average human.
Ha, that’s too funny: the average human composing speech or writing has a lot incommon with the salad chef and the chatbot: none are trying to convey a concept; all are just attempting to put together a set of ingredients that will generate a pleasing sensation for the consumers of their work.
I think the point is not so much about originality (frankly, who cares whether the greatest new invention is termed a new idea or a new combination of old ideas), but that new ideas are not conceived in a vacuum. For example, we give credit to Newton and Leibniz for the invention of integration, but even these two brilliant mathematicians were standing on the shoulders of giants. To take an idea and say ‘that was all me’ is just absurd and denies the environment that made the idea possible. I think this is also relevant in the context of copyright, as this quote was taken from a discussion about copyright.
Raphael:
Ideas are conceived—or, at least, they are developed—in a vacuum! I get some of my best ideas while vacuuming. I think it’s the combination of the routine and non-stressful nature of the task, along with the soothing white noise coming from the machine. Back in the days that you needed to plug the machine into the wall, it was another story, because I always had to allocate some conscious thought to where I was in the room, whether I was about to run over the cord, etc. But with these new cordless models, my thoughts can flow freely, in a way that does not happen if I’m working in my notebook or at the computer, as now.
Roomba sentience: proven.
I also get better ideas the minute I step away from the computer.
I like the imagery of the coloured glass and the kaleidoscope. Even in Twain’s aphorism, however, there is a little rhetorical sleight of hand going on. For surely, granting the assumption that the atomic units of thought are unchanging through the ages, ideas are almost invariably composites. And Twain acknowledges that combinations can be novel. The invention of the bicycle isn’t rendered moot by the pre-existence of the wheel, after all.
Beyond the analogy to human writing and creativity, the quote brings to mind the old philosophical debate between nativists and empiricists. According to the nativists, at least some constituents of human knowledge are innate (either a gift from God or a sophisticated evolutionary endowment, depending on one’s proclivities). For the empiricists, all knowledge is derived from experience. Such debates have been relitigated across domains with language being a central battleground. Pace Chomsky (who insists LLMs have nothing to contribute to science), the current moment in deep learning, both LLMs and ANNs implementing other perceptual and cognitive capacities, gives us that rare but wonderful opportunity to convert age-old philosophical questions into empirical ones.