Goals, Games, and Stories

Writing a blog post on Microsoft Word makes me feel like I’m writing a paper of some sort, so all of this formal language keeps coming into my head. Even writing this sentence in MSWord is a bit weird, but if I don’t remind myself that this is for the blog, I’ll continue to write as if I had some thesis to prove (laugh track.)

But anyway, I was going to talk about The Sims. I was having a conversation with a friend the other day and I think that he hit the nail on the head when he complained that it was Taylorist. I’m not the expert on Taylorism, but it’s generally a philosophy that dominated workplaces a few decades ago by emphasizing repeated motions and compartmentalized processes that maximize efficiency at a given task. Think McDonalds—its “killer feature” being that it made food much more efficiently by using an assembly-line process in which every person does a single task. Yum!

Now, in particular, it went across that same line of reasoning a lot of us may have about the game—we obsess over making our characters eat breakfast and go to work, forgetting that we’re asking them to do the most mundane things that we’re generally avoiding by playing video games. Of course the catch is that we generally don’t operate the same way in real life—eating breakfast, going to work and engaging in seemingly rote self-improvement are generally situated in something other than a single-minded numbers game. I admit, however, I’ve fallen into that trap, and I think many others have; where the idea of self-improvement escapes any sort of idea of life and mindfulness and we suddenly hope to sit down for a few hours and mindlessly repeat some rote task while increasing some theoretical gauge or meter. And, yes, it’s a horrible way to think about life.

And this is where I think The Sims truly fell short of being a great game despite its being one of the most popular of all time. By the way, I’m only talking about the first one, I haven’t played the sequels. But I digress; a truly great game constantly demands that we make decisions of strategic depth—this means that the challenges are always somewhat novel and that there are few (if any) stretches of performing the same task repeatedly. Nor is it even that such repetition makes a game potentially boring, but rather the horrible serotonin trap that can come out of it. There are those times where I can keep getting a little kick of happiness from repeating some action and getting some minimal amount of feedback. When I give one of my Sims a promotion at work or an extra skill-point, there is a slight jolt of satisfaction that reinforces the sense of reward that will come when I pass the next tiny milestone.

What’s important to me about this dichotomy between the rote and the novel isn’t so much what it says for creating a game with great gameplay, but what it means for creating a game that offers a true narrative experience. As I’ve said before (though maybe not on this particular blog), what separates games from other mass media (if you consider games a mass medium—hard to say when such a popular thing is only taken seriously by a narrow demographic) is that it doesn’t offer any sort of real storytelling. Yes, there are backstories, cutscenes, characters, allusions and motifs within games, but most of these things are cosmetic and not at the core of the experience; even the occasional deeper theme that pervades gameplay only scratches the surface of what a real story will deeply penetrate.

As I write this I am in the passenger seat of a car, riding down a stretch of highway connecting Boston to New York, anticipating the sights of big buildings upon my eventual entry into the FDR drive. I’m listening to “Kids” by Kelli Ali—a song with lyrics that semantically mean little more than noise to my rational mind—but the ambience and percussion behind the vocalist’s low and breath-like voice brings about a feeling of mystery. It is the opposite of what Taylorism, rationalism, utilitarianism, classical economics, score cards and every other measurement that’s been mistaken for reality stands for. The soundtrack to a movie that doesn’t exist—but one that I wouldn’t want to exist anyway; a movie that maybe one day I can borrow a few scenes from but leave the rest in the unprocessed ether of the unstructured and the uninterpreted.

This is, of course, the instinct that art and storytelling invoke in us. Nobody gives a shit how many midichlorians Luke Skywalker gains between A New Hope and Return of the Jedi and when I watch Mad Men, I’m hardly thinking about Peggy Olsen’s iron march of progress (though it goes without saying that I’m rooting for her—she’s one of the only decent people at the agency.) But they do have something in common with even the most stereotypical video games, they have goals; that is, goals of a certain nature. We see change over time, infer motives in characters and follow a journey. It’s just that this journey is of an entirely different nature than that which is stamped, filed, indexed, briefed, de-briefed and numbered; it’s one in which the path matters more than the destination—what more mathematical types would call “path dependent.” At its core, it refuses to acknowledge gain without loss, pleasure without pain, joy without melancholy—a far cry from the one-track “pursuit of happiness” (as defined by what the game tells you it is) that The Sims beckons us to.

This isn’t to say that games never offer anything similar to this experience. We can satisfy the simple desire to see interesting things happen by turning up the difficulty a notch higher than we usually would, modifying the game, or (a la The Sims) treating the game like a dollhouse in which we create dramas. A medium is a means of transportation: sound through air, a signal through a telephone wire, meaning through language, a story through a novel, a vision through a painting; so at its core, this “dollhouse” idea is how the process of storytelling operates on some very deep and fundamental level—the reflection of our mind, our environment and our shared cultural discourse off a medium.

But the tools for this kind of exploration are currently insufficient for an experience that even comes close to what I can experience by turning on a movie, opening a book or listening to a song. I want to imagine a game in which people can experience this exploratory mode of play with the full spectrum of behaviors, desires, motives, personalities, webs of causality, empathy, mystery and dramatic weight that we see in stories. Perhaps game isn’t the right word—“game” in the more academic sense of the word meaning known probabilities and mathematically defined sets of outcomes. Whatever I may want to call them, it’s time to embrace the notion of goals, motives and desires that can be understood but not classified—they need to be open-ended; a catalyst rather than some suffocating essence that defines the experience. Motives are as fundamental to both stories and games as symbols are to both narratives and maps; a narrative being an open-ended set of ideas and symbols rather than a strictly delimited abstraction.

Parts of this post, such as those bits about exploration vs. destination, felt so obvious and banal, but that isn’t what’s most important to take away. What’s important is to make this distinction in a way that doesn’t throw the baby out with the bathwater. Goals, motives and desires are at the core of both the most rote gameplay and the most beautiful poetic landscapes—but in making a new kind of “game”, we must embrace it in a different way while still preserving the core ideas behind gameplay. There really is an area between story and game.

Paradigms vs. Narratives

I think I have a simple definition of a narrative:

A paradigm is a sign system that is closed.

A narrative is a sign system that is open.

This also easily explains the difference between game design and interactive storytelling (to an extent; you can think of it as a spectrum, chess is on the "game" extreme whereas something like SimCity or Final Fantasy is a lot closer to the middle. Nothing yet on the "interactive storytelling" extreme.)

I'll elaborate more on this soon and also talk about how this is shaping my work.

For that matter I decided I'm going to also be a lot more open about my work. Scrooges go nowhere.

A definition of narrative?

So many things I wanted to post about, so many work related distractions and so many excuses. Here's one for brevity.

My structural definition of a narrative:

A collection of events* that makes sense of itself without the use of axioms.


*Was hesitant about using the term "event", but I think that it's more precise than saying "signs"--there's a self similarity here I think; the idea of an event is itself inscribed in narrative.


Thoughts?

The Triad

1) When I was in high school I was already thinking about semiotics, albeit without any knowledge of that word. One day in high school, my biology class took a trip to this art activity held by a resident artist who had us all draw "three things that we were made of." A lot of people drew funny and goofy things, I think I took it too seriously. My three things were zeroes, ones and eyes. Zeroes and ones because those are the only two ingredients you need to create information. Why eyes? What is information without something to observe it?

This is still a haunting question in semantic and semiotic theory. That said, two might be the number of symbols (difference), but three is the number of signs. Another reason that I think three is the key number is because signs are not about one-to-one relationships between terms (that's a symbol!) Signs are about multiplicity and ambiguity; a sign has connotations and ambiguities, it can be used not only to name but to give hints and even lies. You need at least three nodes in a graph (in layman's terms, network) to create something more than a simple one-to-one mapping.

2) Speaking of signs and triads, here's a quote that I found from The Name of the Rose that seems to implicitly talk about the triad of signifying/inferring/lying that I talked about earlier:

"It is of use to me as Venantius's prints in the snow were of use after he was dragged to the pigs' tub. The unicorn of the books is like a print. If the print exists, there must have existed something whose print it is. ...The idea is sign of things and the image is sign of the idea, sign of a sign. But from the image I reconstruct, if not hte body, the idea that others had of it." -William of Baskerville

Note: It's always a good idea to take the precaution of saying "biological" instead of "natural" in order to throw smartasses off your trail.

The Garden of Platonic Forms

More on this later perhaps; time permitting, but I may have stumbled upon the fundamental essence of post-structuralism:

Literature has no platonic forms, invalidating the underlying assumption since Aristotle that literature is a way to bring us closer to Platonic forms (Plato believed that poetry and drama was "thrice removed from nature", meaning that it was an imitation of an imitation; but Aristotle believed in the power of poetry/drama to bypass the limitations of Earthly manifestations.)

Structuralism was the most rigorous* way of doing this, but eventually saw its own limits; thus becoming a study of this very cycle of interpreting and then finding new meaning by figuring out how the interpretation contradicted itself or simply didn't apply. Perhaps that's why I have as hard a time as the French understanding any real difference between structuralism and post-structuralism.

--------------------------------------------

*Please take "rigorous with a grain of salt.

Intuition and Mathematics: A Defense

Back in my days at Oberlin College, I had a memorably good math professor who I consider to be one of my mentors. Her grasp of the subject and the toughness of her problem sets helped me learn math that I hadn't thought myself capable of learning at the time. There was one point she made, however, that in hindsight I can't help but adamantly disagree with.

She said that intuition was what messed mathematicians up. Her points leading up to the conclusion were valid; mathematics is a language—you have a set of symbols and a set of rules for manipulating those symbols; and to prove something you follow those rules precisely, moving the symbols into the configuration that you'd like. Intuition can fail; but math never lies.

This fact about math has given way to some amazing tools; not least the ability for a computer to prove new theorems by arbitrarily moving around symbols until they find something of note. Truth is, however, that, as far as I know, these automated theorem solvers have proven very little of value compared to the work of mathematicians; most applications of computers to theorem solving are programs that are custom written to deal with the details of making a particular point such as the proof of the Four Color Theorem. But I digress.

My professor was certainly right that intuition is flawed whereas following rules doesn't fail; but what she didn't acknowledge was that our intuition can always be verified by checking the logic of our hunch; we can make a guess as to what the nature of a proof might be; following which we write up a proof to see if it logically works. In fact, I think this process of guessing and verifying is essential to solving problems in mathematics.

This point can be best made by a certain concept known as computational complexity (if you're not a technical person, please don't panic! It'll be clear in a minute.) Computational complexity is a measure of how difficult it is for a computer to solve a problem; different problems are organized into different levels of computational complexity based on how hard they are to solve. I'd like to focus on two, known as P and NP. P stands for polynomial—the problem can be solved by an algorithm in a polynomial number of steps. NP stands for nondeterministic-polynomial—the problem can't necessarily be solved in a polynomial number of steps, but if you were to guess the answer to the problem, you could see whether your guess was right in a polynomial number of steps.

Now, to be fair, it might be the case that there is no problem in NP that is not in P (but it's highly suspected that this isn't the case.) Assuming that this isn't the case, however, then there are a whole slew of interesting problems that can't be easily solved in a reasonable amount of time, but could be proven were somebody to make a "lucky guess."

But how "lucky" is that guess? Our intuitions solve extremely complicated problems with a very high accuracy rate. They don't prove anything, but they make very good guesses with a very high chance of success in a very short amount of time. After all, our brains are some of the most complex computers imaginable, with complex networks made up of trillions of neurons and connections; we can recognize faces, navigate social situations with countless unwritten rules and outsmart a computer at just about anything that's not a game with transparent rules (and computers still can't play Go for shit.)

Our intuition allows us to stumble upon the answer to a problem with a very good chance of success. After that, we can use logic to see if we've done it right—as the P/NP distinction illustrates, verification isn't nearly as costly (in most cases) as solving.

[End of neat, pre-packaged entry; on to messy stuff]

(Warning: I am less sure about the following. Feel free to correct me on this; and as always, argue with me.) Another way of thinking about this is that with each level of computational complexity, the amount of solutions to the problem becomes increasingly non-linear. Logic can only go so fast (at a linear speed, I suspect) and can't keep up with the increasing complexity of problems. Our intuition has two particular strengths:

1) It runs in parallel, which allows us to detect patterns very fast. Our neurons don't even fire all that fast—their effectiveness is in that we can distinguish information using space rather than time. No computer touches us in that regard

2) It is built to approximate and work in probabilities rather than to think of things in absolutes, which is what my professor's argument encouraged mathematicians to do. Approximations don't always solve the problem, but in a lot of cases a small decrease in accuracy (say, a 10% chance you're wrong), can lead to being able to find an answer much more efficiently. So long as we can verify the answer in an efficient amount of time (that's what proofs are for), then mathematicians can have their cake and eat it too.

The Poverty of Online Information

This is a post I wrote for another website, Good to Know, Inc., but I wanted to share it on this blog in its unedited form (since they're likely to make edits to my writing.) Here it is:

Also, I have no idea why the font is being so screwy.

*****************************************************************************

There are three concepts that, when one looks closely enough, can be seen as a single indivisible triad: narratives, context and memory. They all fundamentally deal with an associative way of thinking and they are noticeably robust: one doesn’t easily forget things (my cousin and I still occasionally debate who was at fault in a scuffle we had when I was seven) and we seem to understand things best when they’re told as a story. They also have another trait in common: they are often lacking if not non-existent when I try to find information from the internet.

Take Wikipedia as an example. It certainly has a wealth of raw information; just about anything one can think of seems to have an article, whether your interest is in hermetian matrices, hardcore bands, Lacanian psychoanalysis or Pokemon trading cards. Even so, I never feel the same sense of fulfillment reading material on Wikipedia that I get from reading a book; I often find myself struggling to walk away with something to hold on to after browsing through this seemingly infinite pool of collective knowledge.

Maybe I’m just of a dying breed of people who hold on to physical books, but more universal reasons seem to be at play. Our brains work in an associative manner; rather than record our memories in a serialized manner like a video camera or a computer’s hard drive, we create memories by connecting various objects, sensations, thoughts, images, sounds and other past experiences. Conversely, we recall memories from the past by seeing, hearing, feeling or thinking something that we’ve associated with it; anyone remember the scene that everybody talks about from Swann’s Way where the narrator takes a bite of a Madeline cookie and his entire past rushes back at him?

A single sensation or reference brings back a rapid cascade of memories and this same phenomenon underlies what might be our memory’s inherent preference for books: we associate its content with the distinct look and feel of the book’s cover, pages, printing, wear and tear. Simply put, you can remember something much better when you attach it to a physical object of some kind.[1] But just as the physical context of knowledge may matter more than we think, so too might the relevance of that knowledge to some larger narrative. The journalist Malcolm Gladwell might oftentimes be criticized for the lack of empirical rigor behind his theses, but I prefer any one of his books to a series of entries on Wikipedia because I believe him to be a masterful storyteller able to weave a larger number of data, anecdotes and ideas into a single story that makes sense of them. Knowledge is much more easily remembered when there is a narrative that brings it all together.

One way of thinking about this is that a narrative introduces redundancy to knowledge. It’s much easier to remember an idea when it’s linked to a number of other ideas as part of a larger theory. If you learn something in a vacuum, there will be nothing to give you any clues to it should you forget it; but by relating it to something else, we have something to remind us of the original idea.

But is it really the case that there are no narratives on Wikipedia? If I look up David Bowie, I’ll find a summary of his life and career and with it there will be hyperlinks to more detailed summaries of particular works or to biographies of his contemporaries; there’s definitely a narrative. Even so, there’s something about Wikipedia that is too summarized and distilled to be the kind of story that sticks well after I’ve finished reading an article. It offers not so much a story as a summary in which details are relegated to the various articles that it links to.

But these links on Wikipedia are inadequate to finish telling the story. They are details that are told in a complete vacuum, unaware of the fact that I came from a page talking about one thing and not another. In Malcolm Gladwell’s Blink, the defeat of the United States Navy 7^th Fleet at the hands of Paul Van Riper’s meager force in the 2002 Millennium Challenge War Games is talked about in the context of how human intuition can be more powerful than the world’s greatest computers, but a Wikipedia article summarizing the thesis of Blink could only tell it in sufficient detail by providing a link to an unrelated page summarizing the incident; one that would fail to have the interpretive richness of Malcolm Gladwell’s own retelling.

A story, as opposed to a simple summary or functional narrative, speaks to us about what is not obvious. Whereas summaries compress, stories interpret by bringing together a wealth of information and constructing a unique idea. Furthermore, through this same process of bricolage, a story is filled with a rich variety of prose and impedimenta that allow us to create a rich network of associations and have an experience that is eminently memorable. A summary, as a simple compression of one or more stories, simply doesn’t have the same informational richness as those stories and as such cannot create the strong mnemonic associations that stories do.

At the end of the day, storytelling is something that is hardwired into us. We comprehend and remember things as stories, not data, and the tradition of storytelling has been a universal ritual as far back as recorded history takes us. This natural affinity of ours towards telling stories can teach us a lot about how we learn and should be taken into consideration as we expand our educational tools into the domains of the internet and cloud computing. The ability to express ideas as dynamic and interactive programs as opposed to static texts and the interconnection of vast amounts of user-created data will doubtless allow us to inform and educate in novel ways; but these innovations must cultivate the human faculty for storytelling if they are to be truly effective as pedagogical tools. As our world becomes increasingly interconnected and hyper-abundant in information, context has become the essence of knowledge; and one cannot understand context without understanding stories.

---

[1] For further reading on this idea, see Nassim Nicholas Taleb’s notebook, Opacity (http://www.fooledbyrandomness.com/notebook.html)

My Interpretation of Inception

I saw Inception a few days ago, and from my collective experience with my other friends who've seen it, it's the kind of movie that you have a strong opinion about; you either love it or you hate it and there's always a reason or five.

I personally loved the movie, I thought that in spite of its flaws that the plot was extremely complex, well-informed and theoretically adventuresome. I saw it as a metaphysical thought experiment that follows in the footsteps of Christopher Nolan's last non-Batman movie, The Prestige. My fascination with the movie came from my seeing the movie as an extremely ambitious thought experiment on narrative, epistemology, language and consciousness; which in itself is funny because it made me realize what a structuralist I've become. I used to always get annoyed when I talked to diehard Freudians because they kept bringing up the same interpretive framework for everything, but it's hit me that whenever I watch a movie or look at a piece of art, I end up seeing it as addressing some question or another about narratives and signs that only a structuralist would really care about.

I'm going to write my interpretation in the form of small vignettes since I don't feel that I could give a single elegant interpretive framework. It's also hit me that I've always worked very hard at linearizing my non-linear way of thinking and that it might be okay for me once in a while to present things in a more free-form fashion that doesn't take up so much energy.

Also, if you don't like things that are "pretentious", then you may not want to read this entry; it attempts to connect a lot of very strange concepts. You've been warned.

*******************************

1) Mal (Marion Cotillard) is quite clearly insane. There is no two ways about it; she's mad. That seems like a rather banal observation, but it is tempting at first to empathize with her after seeing so many layers of dreaming throughout the movie and actually believe that she has a valid reason to be skeptical. But when it comes down to it, Mal has no basis for believing that she's in a dream when she kills herself. She can say "Well, dreams seem real, so there's no way I can prove that this isn't a dream", but she would have to ask that on ANY level of consciousness; in other words, if she was right that she was in a dream, then she would still have to ask the same question when she woke up in the next level up.

2) So there's an epistemological question that goes on with these dreams. You can't know whether you're in a dream if there's no reference to the waking world. This is equivalent to Bertrand Russel's problem of meta-languages. You cannot verify the statement "this statement is true" on its own merits. There needs to be some metalanguage that can say whether that statement is true or false and that can directly compare two statements. If anyone's ever told you that you're comparing apples and oranges, it means that you're dealing with two or more objects that do not have a shared system of comparison, i.e. a metalanguage. Of course, if we constantly concerned ourselves with trying to find a transcendent metalanguage (or for all of you postmodernists out there, a transcendent signified), we'd never get anywhere. Narratives in general, especially religious narratives, help us deal with this kind of problem; from the common ground of shared narratives, we reach some sort of understanding of the world through dialogue.

Madness is epistemic nihilism, a belief that nothing is sacred (which may yet explain people's reactions to my sense of humor). In my last entry, I pondered about how cognitive dissonance allows us to participate in a collective narrative. In Luigi Pirandello's Henry IV, the main character, a disgruntled and possibly schizophrenic impersonator of the German monarch Henry IV raves about the freedom he's gained from dropping the mask and living in a state of constant flux without worrying about contradicting any narrative he or others may have made about himself. He's become quite literally anti-social, a sociopath. On an interesting digression, he is faced with the choice, after murdering his ex-wife's fiancee, whether to keep his mask as Henry IV and trap himself in the narrative of madness, or face the consequences of his crime the minute he stops acting like the supposed lunatic who murdered him; a Madman suddenly forced to sanely masquerade as a madman.

Mal is an epistemic nihilist, and she faces the worst possible consequence for it.

3) The way in which dreams operate is based on a phenomenological theory of storytelling in which the reader's consciousness fills in the semantic gaps left by the text. If I say "He walked down the street" you have to come up with your own idea of what that person looks like, what kind of street they walk down and perhaps how they might be walking. In fact, maybe you don't even think about some of those things until I point them out, or until I give a second phrase that reads "and people laughed at the way he walked." Of course, now that I told you that he was laughed at, you accommodate this whole sentence by giving him a funny way of walking.

In Inception, the shared dream is populated with an arbitrary landscape, perhaps a city, a tundra or a hotel. There are people walking around and perhaps even some familiar locations, but for the most part, these are all arbitrary landmarks. The dreamer populates the dream using his subconscious, filling in for what's left blank with their own personal thoughts. The initial setup of objects demands that the dreamer forms some causality between initial objects in the dream and, as a result, the dreamer creates a narrative out of the objects that reveals information about themselves. This is how the process of extraction works.

4) The process of extraction mirrors how we use language and narratives. Words on their own are arbitrary, meaningless symbols; mere parts of a (very very messy) syntax. By drawing connections between the words that somebody says, that person can communicate with us by allowing us to draw connections between the words and extracting information from the context that emerges when multiple words or sentences are arranged in a particular way.

Narratives work in the same fashion. The "dream defenses" in inception are a perfect example. I personally thought that the dreams having some literal military looked silly, as many people may have. Despite the fact that we know that it wouldn't really look like that in real life, we understood what the military men signified by seeing them try to kill any and all intruders. Every metaphor requires at least two juxtaposed symbols; they are an invariant representation of a pattern. By juxtaposing two or more symbols, you can imply some sort of causality or context that mirrors something that people understand and therefore turn meaningless symbols (such as say... a bunch of talking farm animals) into something that helps us understand the world (such as, just for the sake of argument... Animal Farm.)

5) Extraction works by framing a person's consciousness with arbitrary symbols and then causing them to involuntarily fill in the blanks with parts of their personal life. We're simply hard-wired to look for patterns and try to make predictions by accounting for the unseen. The dreamer reveals their secrets by allowing the intruders to see how they connect the dots and account for what isn't present in the dream.

In other words, it is the context of what somebody is saying that reveals the truth. Otherwise, the person could be lying or telling the truth or just saying nonsense; but if we compare it to what they've said before and under what circumstances, then we can assign some sort of information content to the words. The logical conclusion of this is that the truth about a person comes from what isn't present in what they say, that it's always looking between the juxtaposition of somebody's words, sentences or actions. This is why our English teachers always tell us to "read between the lines."

6) As an extension of (3), art and religion make for very good learning tools because they utilize our brain's incredible talent for storing patterns as invariant representations. Stories represent a wealth of highly abstracted shared wisdom accumulated over the course of civilization; all in a convenient narrative form. But I digress.

7) The dreamscapes created by Ellen Paige use logical paradoxes in order to "close the loop" of the dream. Most notable was the infinite stairs reminiscent of M.C. Escher. I don't fully know what to make of it, but such paradoxes seem to tie in well with Bertrand Russel's metalanguage problem. The dreams have within them some sort of "incompleteness" in that the dream simply cannot account for all possible questions the dreamer might ask. By making the dream physically loop on itself in a seamless fashion, Ellen Paige can make sure that the dreamer will simply not be able to answer certain questions such as "does this dream end or is it infinite?"

In real life, we're faced with a similar phenomenon. Many logical paradoxes exist within our world, one of which has shown that some true statements in mathematics simply cannot be proven. This is called Godel's Incompleteness Theorem. In this sense, the dream is arguably just as real as "real life" because it has a set of symbols upon which we are able to endow a narrative and has logical paradoxes which shroud certain epistemic questions about reality.

8) And logical paradoxes might just be the most essential thing to the fabric of reality. If we had "total knowledge" by which all things were proven with a consistent set of axioms, then we would no longer have any sort of experience because there would be nothing left to learn. All experience is caused by the activity of the brain making sense of its inputs and all narrative is us drawing causal links out of an infinite amount of possibilities to account for some series of events (or more accurately, some set of symbols.) The existence of epistemic limits means that we continue in the activity of learning and creating narratives.

In my opinion, narratives are the essence of experience because they're about imagining what isn't there; we live ineffably. Things continue to be ineffable in the absence of irrefutable proof. But perhaps it's also because reality works just like the dreams featured in Inception; that the universe just doesn't have enough material to answer all of our questions and so reality closes itself under a logical paradox that leaves us to continually guess with narratives. The universe is hardly the static, sound and complete ontological entity that the Enlightenment envisioned, it's a constantly fluctuating narrative where ideas like "true" and "false" are far less understandable than we think.

And I think that's why Christopher Nolan made the last scene such a cliffhanger. To be honest, I actually hated that scene, I think it was an extremely un-subtle way to address a question that the movie already implied and I believe that by putting it in the forefront, it made people take the problem so literally without wondering whether it even mattered whether Leonardo DiCaprio was still dreaming.