On Lore.

“A Narrative Designer's job is to answer the questions before the player can ask them.” - Edwin Mcrae, lead narrative designer on Path of Exile

“The story is awful at best, its [sic] going to be just as bad in POE 2.0. Nobody cares about the story, except a handful of lore dudes.” - deleted reddit user, PoE subreddit.

McRae wrote that narrative in games has three purposes, objective-granting, guidance-giving, and meaning-imbuing. I would quibble with this on a number of fronts. Firstly, the objective-granting purpose is simultaneously too narrow (merely the announcing of a goals existence to the player) and redundant (a goal consists of meaning and method, captured by the other two purposes). If one wanted to define the difference, a two way split between sub and superstructure might be more clean. Substructure here means narrative as a tool for organising, explaining, and understanding gameplay actions. Superstructure meaning game design is a tool for making narrative legible, resonant, and compelling.

A different angle might be to define narrative by the level of abstraction the game assumes, allows, or is experienced on. An example of what this means is how players describe their own actions during or after gaming. Compare the following statements:

“I’m going to click the mouse” “I’m going to use my spells on that guy” “I’m going to take revenge on that soldier for killing my companion!”

The first level is nearly never mentioned - most players understand and are comfortable with the idea that a game provides a level of abstraction away from being at the computer. If narrative is that which is experienced, then narrative carries implicitly with it the notion of controlling this level of abstraction. When players feel dissonance with the narrative, it is likely that the assumed level of abstraction is in conflict with the player’s preferred level.

How then, does lore fit in with this framework of narrative? A starting definition of lore is that it is a collection of facts (F1, F2, …) about the gameworld. Often, there is another component to the definition: lore is the sub-collection of facts that deal with the gameworld to the exclusion of narrative. There is often a pejorative sense with which the word is used, as in, “the story is just lore” (i.e., the narrative is a collection of facts presented to the player, with no attempt to connect it to the abstraction layer below). Even for lore that is considered high quality, oftentimes it is presented and consumed as a separate product or narrative, as a way to spend more time adjacent to the gameworld that they feel affection for or connection with. The ‘lore video’ phenomenon, where a content creator digests and presents lore in an easily understood format, points strongly to this ‘adjacency’ being the case.

The ur-game for lore videos, and for a very specific definition of lore is, of course, Dark Souls. The narrative of the game was of struggle and overwhelming odds in combat. To support a non-linear, staggered progression of power, a non-linear, staggered procession of information about the world was used. Facts distributed via item descriptions etc elided the common deployment of lore as a narrative crutch - to imaginatively suggest, Dark Souls leveraged marginality. This had the obvious downside of having the narrative rely on a lower level of abstraction in general. The reason why Dark Souls succeeded in this as opposed to many of its imitators is that the narrative generated via gameplay supports this wilful obscurity and is enhanced by it in a symbiotic manner.

Is the way a boss moves, lore? It is caught between a fact and an action in the gameworld, or reaction. Are character build restrictions lore? It is a preclusion of actions justified mechanically as well as with reference to gameworld facts. Let’s separate out game actions (A1, A2, …) from facts. Facts are static and so reference only the gameworld. Actions are dynamic, i.e., they are necessarily player-referential. I am taking game action to be more broad than a game ‘choice’: accepting the tutorial quest from Tarkleigh is a game action as it is player-directed. So now the boss moving like a horse is a fact, the horse-like movement in response to the player is an action. Both facts and actions can be narrative.

Let us return to McRae. The quote at the start of this essay implies that a successful narrative is adept at creating coherency and legibility. This means that the gameworld has principles, and those principles can be reasoned about via the visible effects on the world to the extent that a player can build intuition. This immediately gives a more appealing purpose to narrative: intuition-building. F1 gives context to A1, or A2 creates F2 (the meaning the fact is a supposition by the player). Narrative then is the relationship between facts and actions, or facts and facts, or actions and actions. We can now formalise and organise McRae’s framework:

• F implies A: Guidance-giving. Context has given the player intuition for gameplay.
• A implies A: Objective-granting. Gameplay has given the p[layer consistency and fluency towards mastery of the game system.
• A implies F: Meaning-building. Gameplay has given the player an intuition for the gameworld.
• F1 + F2 implies F*3: Lore. This is the easiest to understand, and the hardest to get players to care about as it is not necessarily player-referential.

This next comparison will sound rude, but I do not mean to belittle or mock by saying the most analogous activity to game narrative is that of accompanying information to toy-sets that provides scenarios for play. The lego set instruction manual is an organisational structure that informs and guides play. The character bios for barbie dolls provide lore, yes, but more importantly they provide an imaginative jumping off point to build intuitions and expectations of play.

One interesting corollary of defining purpose of narrative this way is that “lore dudes” now have claim to a narrative, albeit a parallel one: the act and ‘play’ of piecing together lore scraps into a coherent work involves a level of intuition-building, about fleshing out an understanding of the gameworld so that you can rotate it to more accurately slot in facts. With a higher level of abstraction than the game, F1 + F2 implies F3 becomes equivalent to a new A. Regardless, this new game is related but distinct to the narrative of the game proper, even if it uses the same pieces. A implies F: Meaning-building. Gameplay has given the player an intuition for the gameworld. Lore is the application of facts about the gameworld that doesn’t alter the memories of game action.

Now let’s situate this discussion in the context of ARPGs. A common refrain amongst players: “I just skip the cutscenes and dialogue.” Lore is an intrusion onto the game from a higher level of gameworld abstraction. It is what one must grit one’s teeth through while mashing skip. Why is this the case? My first instinct is to reject this as chauvinism - the archetypical player loathes anything that slows down ‘gameplay’, to the detriment of all other signals, and feels suspicious of narrative as a concept, too lumberly and awkward, too old fashioned, the wrong medium, not what I bought the game for. I think this instinct is unfairly reductive. What players are expressing isn’t that they disagree with narrative per se - they will delineate between story and gameplay while at the same time admitting into gameplay various concepts associated with narrative - mood, tone, consequence, how action makes them feel, how power in game feels rewarding. The latter is generated and supported by gameplay verbs, character build, items, loot, etc, but these are always within the context of the gameworld. “I don't know how this person will act, but i can make a good guess” Where shall I get weapons? Or magical items? Here comes a grizzled man and a woman in a gown (discussion of narrative shorthand and the function of tropes and stereotypes, and the effects thereof, is left to the reader).

Good narrative is not necessarily frictionless. The process of intuition building is a negotiation between rewarding the act of asking questions, and challenging the simplest answers to those player-posed questions. It’s here that I most seriously disagree with McRae about the function of narrative design. While I do think it is important to build intuition, I disagree that the purpose of said intuition is to answer questions, as if the player is wanting to ‘arrange proper nouns’ as Noah Caldwell-Gervais puts it. No, the intuition building itself is the critical aspect. Answers to narrative questions are useful and satisfying insofar as they support the process of imaginative participation. When facts of the gameworld don’t aid in this participation, they become lore in the pejorative sense. Narrative is not just supported by linking context and action, it is inherently connective tissue.

To quests, then. The quest is a narrative structure made manifest as game action. While I maintain that narrative is connective, the quest is a pre-connected and predetermined body of facts and actions. To illustrate what is required for a successful quest, it helps to start with the grandfather of unsuccessful quest design - the fetch quest. What is objectionable about this structure?

Quest granted - Movement 1 - item acquirement - Movement 1 in reverse - Quest completed.

The first problem we can notice is that the last part of the structure is, in the base form, repeated content. Secondly, if the act of movement is routine or designed to be unobtrusive, the movement sections are simply down time or ‘busy-work’. Thirdly, the item acquired has meaning only insofar as it relates to the quest (a ‘macguffin’). Put another way - the quest lacks context and connectivity to other parts. Lastly, the act of quest granting and quest completion is static. This makes the quest grant F implies A, except that instead of intuition-building the objective granted is isolated. The sole reason for the quest is that there is a quest. Even before spicing up other sections, the effect of this last problem is severe. Contrast the fetch quest with a quest of the same structure but ‘unofficial’, i.e., the player has formed the intuition about the gameworld that delivering item A to person B gives interaction C. Because this narrative is now connected to the player’s intuition, they are reasoning about the gameworld and thus imaginatively participating. This immediately makes the quest structure more palatable, as now the F implies A starting relationship is narratively charged. Regardless of the outcome of the ‘quest’, the player will add to their understanding of the gameworld.