Sponsored By

The Last Guardian is a Masterpiece

Meandering thoughts about a flawed modern masterpiece.

Josh Foreman, Blogger

December 29, 2016

19 Min Read



Here’s a game made by my favorite designer, Fumito Ueda, 11 years in the making.  He made (It’s stupid to say any ONE person “made” a game since there’s usually a giant team and most of them are indispensable in making it happen, but that’s the parlance.) some of my favorite and very seminal, influential games: ICO and Shadow of the Colossus.  These games were groundbreaking for several reasons.  One reason was something called “design by subtraction”, nicely articulated here.




In the early 2000’s console games were getting more and more complex, demanding and using all the buttons on the controller, competing with each other to be the game with the most features and mechanics.  Ueda, an art student with no game design history came onto the scene and wanted to do something different.  Instead of breadth, he went for depth.  He decided on a core mechanic that was both unique, and quietly subversive.  I’d go so far as to call it revolutionary.  In a market full of gruff men shooting at monsters, or wacky/cool anthropomorphic mascot characters spinning and jumping on monster heads, Ueda gave us a game about holding hands.  ICO is still structured like most games.  You play a protagonist in a conflict.  You traverse an environment to resolve it.  But within the bounds of this sandbox, Fumito’s toys were very different than the other kid’s toys.  There are monsters, yes.  You can’t kill them.  You are almost powerless against the smoky (that is REALLY not how smoky should be spelled!) apparitions.  When they come, they come for a mysterious girl you’ve freed from a cage in the massive labyrinthine castle the game is set in.  They try to pull her into a pool of shadow they emerge from.  You get a stick.  But you have something far more powerful.  A hand to hold.  


The verbs (game mechanics) that games gave you back before the indie revolution of the mid-late 2000’s were mind-numbingly small.  Jump, shoot, punch, flip, reload, block, shoot-some-more.  Imagine the shock to the system caused by a new verb entering that lexicon called… hold-hand.  Suddenly your paradigm of aggressive-forward-momentum-towards-destroying-enemies is inverted.  Now you’re tenderly moving back to grab the hand of a friend.  This was the beginning of a revolution.  The headspace you occupy while playing ICO is usually relational, not confrontational and violent.  As you lead and protect the mysterious, passive girl you can’t help but form a bond with her.  You save the game by falling asleep on a couch with her.  You get nervous the further a traversal puzzle forces you to walk away from her.  Your success in the game was tied to her survival, not an arbitrary health bar or number of enemies killed.


Then, in 2005 Ueda followed up that masterpiece with another called Shadow of the Colossus.  Bigger and more ambitious, but still building on the inverted paradigm that ICO offered.  In this case, the contrast with the rest of the industry that stands out the most is the narrative delivery.  It’s minimalistic in a time of bloviated cutscenes.  It’s vague and subtle in a time of over-emphasised-technobabble and hit-you-over-the-head-dumb-Hollywood-wannabe scripts.  And it took the hero paradigm almost every game dealt with and made you question the assumptions that the game medium had always leaned on as a crutch.  (Mainly, the implicit assumption that the player/character must be ‘the good guy’)  All of this, and again, a jaw dropping new core mechanic.  Climbing.  Climbing had been in games before.  Never climbing on building-sized, moving, living creatures.  This was both an aesthetic and technical marvel at that time.  It’s only even been attempted in games a handful of times in the 11 years since then.  It’s REALLY hard to do, from every angle.  Technical, design and art all have to be absolutely brilliant to pull it off.  Team ICO was absolutely brilliant.  


With that background established, I want to talk about some of what The Last Guardian means, both to me, and to the medium of games as a whole.  I’m going to start with the common criticisms that I’ve seen.  There are three main complaints.


First, the controls, camera and physics are really squirrely.  It’s just really hard to control the boy, too easy to fall in many places, and impossible to fall in others.  Many puzzles rely on the physics working right, and since they usually don’t, there’s a lot of trial and error and luck involved with progression.


Second, the AI of your animal companion, Trico, is frustrating.  Many puzzles rely on Trico following your… suggestions.  And, like a real animal, it seems to usually just be confused by what you’re asking it to do.


Third, and I think most interesting and full of implications for the artform of games, is the idea that the first two issues would not be issues if the game had come out a handful of years after Shadow of the Colossus like it was meant to.  (Around 2007 - 2009) It didn’t.  It went through development hell for at least 11 years.  Here’s a brief history, but it doesn’t mention the fact that Ueda said in an interview that work had actually started on it before SOTC had even finished, which means it really started in 2005.  http://kotaku.com/5964547/kotaku-timeline-the-last-guardian


This criticism about the ever-changing standards of modern artforms has always eaten at my soul.  It really bothers me that people can’t create a truly timeless film or video game.  And to a more limited extent, even a novel.  Whereas we can appreciate paintings, pottery and sculpture from thousands of years ago.  (Not to mention physical games like chess or dice and cards)  I want to know why this is.  Why is a film like Citizen Kane a ‘classic’ and a ‘masterpiece’, yet it’s unwatchable to most modern viewers?  There’s a lot of specific answers to this.  Modern sensibilities, pacing, acting styles, etc.  But WHY do those sensibilities change?  Old arcade games are very similar.  Classics like Pacman and Donkey Kong can’t hold my attention for more than a couple minutes.  I … HATE this.  I hate that the art we developers pour into our work is so ephemeral.  It’s like writing for a newspaper or something.  Pointless.  


I mean, it’s not.  But it FEELS pointless to ME.  Probably has something to do with a primal subconscious desire to escape mortality.  If I’m going to pour my heart and soul into art, I want the biggest possible impact, which means the biggest possible audience, which means as many people from the future as possible.  This sounds like a personal problem.  There’s certainly no solution that I can imagine.  You can’t stop the changes in cultural tastes and expectations.  It just takes the wind out of my sails a bit.  I mean, I am still driven to create art because that gives me joy.  It is only diminished by the fact that its impact is ephemeral.  And I suppose I ought to just come to grips with the fact that everything all humans do is ephemeral… and get over it.


And THAT’S my review of The Last Guardian!  Hope you liked it!  


Juuuuust kidding.  


Let me tell you what is brilliant about The Last Guardian, and I why think it deserves to be considered a masterpiece despite the truth of the criticisms.  First, let me get one thing out of the way.  I didn’t have much fun playing this game.  For the uninitiated, the idea of playing a game that isn’t fun sounds bewildering.  Saying a game that I didn’t have fun with a masterpiece of a game is doubly so.  That’s because you think games are toys.  Some games are toys, and I love them.  Some games are not.  Just like a film can be not-fun, yet still a masterpiece, so it is with games.  However, I don’t think this game was INTENDED to be not-fun.  It’s just that the primary mechanics revolve around things I don’t particularly enjoy.  Progression is almost completely environmental-puzzle-based.  I’m bad at environmental puzzles.  I don’t enjoy them.  Though I LOVE making them.  Weird, right?  Anyway, the fact that this game is mostly a series of puzzles that rely on sketchy controls, physics, and unclear AI makes the moment-to-moment experience pretty bad for me.  There were many times where the solution seemed clear, but the randomness of the physics objects you have to interact with, vagary in what is and is not climbable, or Trico not cooperating, made me seek out other solutions for what felt like interminable amounts of time before finally coming back to my first impulse and finding that stuff just happens to work right this time.  I can almost guarantee you that this convoluted soup of problems is what took the game so long to make. (Along with Trico, which I’ll get to later.)


If all of those systems worked together flawlessly, the moment-to-moment game would have probably been… fine for me.  Probably still not really FUN, but fine.  This is exactly like what ICO did.  It was also mostly environmental traversal puzzles.  But they weren’t as dependant on wonky, semi-broken systems.  


So WHY do I love this game so much?  Two reasons.  First, it’s for what it attempts.  Second for the a long list of subtleties that demonstrate how execution focused around a central idea can create a new and powerful experience.  


So what does this game attempt?  Well, it takes the companionship-forming mechanics from ICO, and the exhilarating experience of climbing on a gigantic beast from Shadow of the Colossus, and puts them in an even more epic environment.  It keeps the minimalist narrative form of the previous games, leaning heavily on environmental storytelling and a handful of voice overs.  All of this is FOR something.  And that something is the experience of bonding with an animal companion.  


A brief aside.  I am a creature accursed by God with allergies.  Not just fur and hair, but animal saliva as well.  Doubly accursed because I LOVE animals. And animals love me.  They always find a way to run up to me and lick any exposed skin, giving me an itchy rash for a couple hours.  I want a little dog so bad.  I had a couple dogs as a kid. My favorite was a French Bulldog named Bilbo.  I miss him.  But I don’t miss itching all the time.


SO, Ueda wants to make a game about forming a bond with an animal.  There are animal games out there.  Tamagotchi, Seaman, Creatures, Petz, Foo Pets, etc.  They all revolve around a simulation of a cute (with the exception of Seaman) creature(s) that you virtually feed, pet, train etc.  There’s usually no goal.  Maybe a competition you can win by training them well or something.  But any emotional investment comes from a gameplay loop of earning things to give to the pet that makes them play an animation.  I don’t mean to be flippant about this.  Any time you break down a gameplay loop it seems facile and pointless.  It’s true of sports too.  Anyway, the thing that sets The Last Guardian a league apart from these kinds of pet sims is that the relational loop takes place in the context of an adventure.  In real life, if you put two dudes together in a waiting room for a couple days, they may learn each other's names and maybe their jobs.  Put them in the trenches of a war and they’ll be bonded for life.  I think this is the phenomenon that the game is leveraging.  It’s not giving you a toy, like every other pet in a game I’ve experienced.  There are games that give you animal companions that do similar things.  Dogmeat from some Fallout games.  That dog in Fable 2 and 3.  Various pets in MMOs.  What these games are attempting is to find those heartstrings by way of trial by fire.  But what set’s TLG apart from those other games is that Ueda trims the fat and FOCUSES on the animal companionship like a laser, organizing all the other elements of the game around this goal.  And, if you can get past the physics, camera and control issues, it succeeds brilliantly.  But why?  How?  That’s the second part I want to get into.  The execution of the concept.


Here’s a list of the great ideas that all worked towards a brilliant execution of Ueda’s goal of giving players the experience of bonding with a giant fantasy animal.


  1. The environment you traverse is an ancient dilapidated city in the bowl of a massive extinct volcano.  As a player you wake up next to this mysterious animal in a mysterious location for a mysterious reason.  The desire to find the answers to this mystery are not separate from Trico; they are enmeshed.  This is good decision number one.  This ensure that Trico is never ancillary to the player’s desires.  Great game narrative design always aligns the real-life player's desires with the plot and core mechanics.

  2. Relationship arc.  In a sparse narrative like this, there’s not many tools for character development, especially with almost no dialog.  Instead, the game focuses on a relational arc.  It starts you out as antagonists.  Trico is chained up and doesn’t trust or like you.  Trico is bigger than two elephants put together and fast as a cat.  He has terrifying glowing eyes.  Strange sounding growls.  Clawed hawk-like feet.  Heavy armor.  Setting the relationship this low at the beginning means all progress is anchored well.

  3. Progress and puzzles are driven through relational growth.  This game’s experience probably couldn’t happen in an open world game like Fallout or Fable.  Because the design-by-subtraction decision was made to focus on the relationship, they have the resources to explore the full range of relational dynamics.  And they created a linear environment to force that growth.  Unlike Fallout, where you are limited to telling your dog to dig, attack or come, TLG revels in exploring every combination of relationship dynamics. This is why, even though I don’t love environmental traversal puzzles, I love this game.  Many of the puzzles require you to learn something new about your companion, trust your companion, protect your companion, need your companion to trust you, to protect you, etc.  I felt fear OF Trico.  I felt fear FOR Trico.  I felt gratitude for Trico.  I felt protective of Trico.  I felt sad, proud, angry, embarrassed (Poor thing get’s it’s head stuck in a small hole trying to follow you!)  I felt frustrated, vindicated, confused, silly, I laughed out loud several times.  I cried real tears.  All of this happening in an artform where the VAST majority of games are content to make me feel: “I’m a badass!”... and that’s… it.

  4. Related to the last point, but worthy of its own point; The fact that the puzzles involve psychological healing of Trico is truly a stroke of genius.  They set up that Trico has been controlled and abused through the use of these stained glass eye insignias.  That provides an excellent chance for the diminutive boy to ‘clear the way’, giving more variety to the puzzles and simultaneously growing the relationship.  Trico simply can’t make itself physically move past those icons.  And so when the time comes that the boy ends up in mortal peril on the other side of one, and Trico summons the courage to overcome that, it truly feels like an earned milestone, and a real stand-up-and-cheer moment.

  5. The size dynamic.  Why didn’t Ueda give us a game where our animal companion is the size of a normal dog or cat?  That would be more relatable, right?  I think there are a lot of reasons.  One is that Trico’s size facilitates large and visually awesome environmental traversal.  For all I know, that could have been the ONLY reason.  However, the decision to make Trico huge gives us something really special.  The fact that Trico is huge, AND that we need to climb on and around its body means that we forced into a close, intimate physicality with Trico.  This cuts back on the ‘giving commands’ dynamic that most game pets fall into where they merely become a mechanism.  It’s harder to feel like Trico is merely a tool for progression when you are physically holding onto its body so much.

  6. Care and feeding.  All pet sims and many games that feature pets have various ways to feed and groom them.  But again, the size dynamic makes this something special.  Feeding, often means getting close to a mouth that can easily consume you.  Grooming Trico requires climbing on and through a forest of feathers.  And the grooming isn’t merely an aesthetic thing for its own sake.  Throughout your adventures Trico is attacked, with spears, teeth, swords, and psychologically.  It’s up to you to pluck the spears from Trico’s hide, to smooth out the ruffled feathers and clean off the blood after such encounters.  And even more touching, you need to comfort your companion, because Trico get’s really upset after a fight.  Here, the voice acting really shines.  The actor doing the boy’s voice is REALLY good.  Even though the words are in a mysterious language, the intent and tone is clear, endearing, cute as hell, and perfect.

  7. It’s been said that voiceover is the laziest problem solving device in film.  I’m someone agnostic on that point, and quite skeptical about applying it to games.  TLG uses it, in my opinion, to great effect.  The main reason I praise this feature is for what it allows them NOT to do.  It allows them to avoid both cutscenes, and/or user interface solutions which would lay bare the mechanical aspects of the relationship dynamics.  For instance, there’s a moment where the voice over comes on (it is presented as a grown up version of the boy speaking his memories) and says something along the lines of “At that point our bond had deepened enough that Trico trusted me enough to follow some of my directions.”  Then a button prompt pops up telling you how you can give commands.  This could have been done with a ‘trust meter’ that you see fill up as you groom and feed Trico or something like that.  VO is a far superior choice.  It also facilitates dummies like me getting through the game by delivering hints if the player is stuck in one place for too long.  

  8. Aesthetics and tech.  The environments are gorgeous.  The lighting in particular is a signature of Team Ico, (now genDESIGN) and creates many momentsof breathtaking sublimity.  It’s got the best ivy I’ve ever seen in a game.  The level design is good.  Not as good as ICO, in that I didn’t have many of those moments where I realize ‘I’ve been to that place way down there!” or “I saw this place way in the distance at the beginning of the game and I’m there now!”  But still beautiful and containing enough surprises to keep you interested.  All of that is great, and a big part of why I’m a game artist and designer. But there’s not much here that hasn’t been done before and better. So that’s not what set’s this game apart.  No, what makes TLG special is Trico.  I don’t think I can overemphasize how breathtaking Trico is as an artifact of human ingenuity and artistry.  Quadrupeds are hard to do in games.  Rigging the animation system and the physics of dynamically planting the feet is hard.  Getting them to look right while traversing uneven terrain is really hard.  Getting them to TURN realistically on uneven terrain is super-really hard.  Getting them to turn realistically on uneven terrain in enclosed spaces… Impossible.  Never been done before.  Now, take THAT, and make that creature adorable, terrifying, loving, fun, soulful… This is an unparalleled achievement that sets a bar that I don’t think will be raised for many many years.  On top of that, give an artificial creatures eyes so real you feel like you’re looking at a real animal.  Give it feathers that blow in the wind, all individually.  Feathers that bunch up against each other as the limbs and neck move.  Feathers that get mussed up from a frakus.  Now give that creature a massive suite of animation and AI systems that totally sell an inner soul.  It’s head and eyes tracking any sudden movement, watches birds and lizards, fixates on food, follows the boy.  Expressions along the whole range from mildly concerned to worried to fear to terror.  Vocal mannerisms that reference many animal sounds we’ve heard before, but are not a direct recording of any particular animal, and yet communicate so well.  Pacing of movement that is dynamically built based on the environment and various stimulus.  Pauses, dashing, frollicking…  If everyone knew how hard all of these things are to pull off, I don’t think ANYONE would be complaining about how long it took them to make this game.


Ultimately, the reason I’m classifying this game as a masterpiece despite it’s flaws come down to the fact that it did what ICO and Shadow of the Colossus did in their times.  It set a new bar for what video games can do.  The fact that a game creator can head three games in a row that do this is… unprecedented.  You might be able to say that about Miyamoto or Sid Meier or Will Wright.  But I think there’s a lot more noise-to-signal ratio in their portfolio.  I’m saddened that so many reviewers are focused on the obvious flaws with the controls and AI, and apparently unable to see just what an achievement this masterpiece is.  I think in time the brilliance will be recognized, but I’m worried it won’t be soon enough to commercially reward that brilliance.  And that’s the biggest crime for the medium of games.  When we don’t reward our true visionaries, we’ll continue to stagnate, and spin our wheels making more games that make us feel like a badass... and nothing more.


Also, because it’s related, here’s a tutorial I made for upgrading the base of the collector’s edition statue.  https://www.youtube.com/watch?v=tyavDidpdyk


Read more about:


About the Author(s)

Daily news, dev blogs, and stories from Game Developer straight to your inbox

You May Also Like