Beyond Default D&D: Worldbuilding Made Better

The release of 5th Edition has both enabled and benefited from a revival in Dungeons & Dragons across the world. On the whole, I think, 5th Edition is a good system, possibly the best D&D system released yet. Certainly it has the smoothest gameplay and is the most accessible and easiest to use. (Disclaimer: I have never played 4th Edition and do not know it well, nor Pathfinder, so those may be competitors for all I know.) Besides the mechanics, my main concern is really with worldbuilding. While roleplaying itself is fun, for me the juice, the real vigour is in the worldbuilding that provides the context for the roleplaying, and this goes especially for D&D given its high fantasy setting. Few things are more reliant on doing well-established tropes well as high fantasy is: after all, being tropey is precisely the point of that genre, and D&D has always recognized this (as does Shadowrun, for that matter).

Given that fact, there are certainly things to be satisfied about from that point of view in 5th Edition. Although it is not new for this edition, getting rid of the bizarre ‘race’ based restrictions on class is a clear step forward in general: no longer can gnomes, for some unaccountable reason, only cast arcane spells as illusionists, as was the case in 2nd Ed. They have also decided to abandon negative racial modifiers, leaving in place the racial bonus system but removing the malus, which strikes perhaps the best balance between the demands of the trope (why else bother with the idea of separate ‘racial types’?) and the understandable desire to not associate the term with negative attributes. One can wonder whether using the term race in this context at all is still appropriate and helpful. ‘Subspecies’ might sound too clinical or biological, but something like ‘physical nature’ or simply ‘character type’ would do just fine, especially given how vague terms like ‘class’ and ‘archetype’ already are anyway. But that is not what I want to talk about here. I have a few enduring frustrations with the worldbuilding assumptions of what I call default D&D, and I want to rant about those here instead.

I should clarify that when I say default D&D, I mean the kind of high fantasy setting that is most commonly and by default the one provided for the D&D mechanics and concepts to operate in. In the current edition, (a version of) the Forgotten Realms campaign setting was fully integrated into D&D as the setting, used as the standard for examples, suggested names and locations, concepts, and so forth. It should be acknowledged, of course, that the 5th Edition books as all D&D books make it clear that a DM should feel free to design their own stuff if they want to, but usually as an afterthought; most of the attention in terms of story and lore, also important for the purposes of illustrating the usage of the mechanics and concepts, goes directly to the official version of this setting. Several editions worth of this practice (meaning at least 15 years of gameplay) has meant that this setting’s assumptions about worldbuilding have powerfully influenced D&D games. Most DMs, I suspect, will use the standard setting or something that hews closely to it, simply to make their life easier. And even those that design an entire world of their own, such as Matthew Mercer has done in his campaign for the hit D&D show Critical Role, still – as his example shows – consciously or unconsciously adopt a great deal of the imaginary that comes with it. Now don’t get me wrong, I am not arguing against high fantasy tropes, D&D classic mechanics, or even high magic settings. But there are a few persistent elements of worldbuilding ideas that I dislike a lot about this default D&D, and I will explain why.


Language. One of the concepts, and accompanying mechanics, carried throughout the default D&D multiverse is an absurd approach to language and language use in a fantasy world. It is bad enough that nomenclature and ad hoc inventions in pulp fantasy have become so notorious that every linguist cringes at the sight of yet another Fantasy Apostrophe. But in D&D, this is taken one step further, one step for the worse. While much of the brilliance and imagination of Tolkien went into his carefully layered historicism, a major vector of this – apparently easily overlooked despite his frequent references to its manifest importance – was his attention to language and linguistics. His invented languages have become justly famous, but not just because of the work he put into them and his professional knowledge of linguistics. More importantly, he spent time thinking about how languages are used, how they evolve, and how people interact with them as representations and expressions of their cultural life-world. This is an important part of what gives Tolkien’s world its element of verisimilitude: when Gandalf has to find the right password at the gate of Moria, this password is one that would actually have been thought of by the elves who wished to pass, and would have been simple enough that the dwarves of Moria could learn it. And it is not an English word, even though Tolkien (in his imagined role as translator) provides us with one.

In D&D, on the other hand, the default worldbuilding would make the most dedicated 19th century romantic nationalist cringe. Available languages invariably, across editions, consist of ‘Common’ plus one (and exactly one!) language per ‘race’. Elves are presumed to speak Elvish, which is a single language, no matter where in the world they are from. This reproduces bizarre racial thinking in fantasy far worse than Tolkien himself ever did. Can one imagine the idea that everyone with a certain hair or skin color would be assumed to speak the same language, whether they were born in Lima or in Tokyo or in the foothills of the Carpathians? That every ginger in the world speaks Redhead as a native language? It’s blatantly absurd and would never happen, and yet in one D&D campaign after another – and I take here the many live streams of D&D now available via Twitch and other media as representative proof of practice – it is absolutely taken for granted that this should be used and makes sense.

Sure, it is mechanically simple. But not really much more so than if one scrapped the idea and simply gave arbitrary names to a couple languages on a first geographic and subsequently cultural-historical basis, as every single natural language ever known to humans has had. Moreover, the DM anyway is still bound to explain what ‘Common’ is – and whose Common anyway? – and where it operates and where one cannot be expected to use it, so that one might as well actually spend some time thinking it through and coming up with a couple languages corresponding to areas and cultures besides that one. I am not saying every DM has to become a dedicated conlanger (although it is fun!), but this is such low-hanging fruit; and yet because of the ridiculous approach of default D&D as expressed in the handbooks, it is rarely thought about or explored. (It may also be that the degree of monolingualism common in Anglo countries makes DMs and players from those places think less about language and how it works in a diverse world, but I don’t know that for sure.)

Another bad result of fantasy convention, one that has fed into the default D&D setting although it is not so explicitly spelled out there, is the total disregard for whose language perspective is being taken. Again, Tolkien invented his notion of himself as the ‘translator’ of a series of narratives, for which he even provided in-world sources and origins (the ‘Red Book of Westmarch’ and so forth). The reason he did this is so that he could render Westron as English, and then have everything relate to Westron as they would appear to the Hobbits: so the Rohirrim have Anglo-Saxon naming themes because their language was a very antiquated form of Westron in the experience of the hobbits, and Old English is the equivalent for modern English speakers.

One does not of course have to create such a construction for a D&D game. But what one should do, to maintain a sense of internal logic, is think about who named things and for whom. If your version of ‘Common’ is generally spoken in the lands in which your party operates (which surely seems a good idea), then it would make sense for most of the names to be in some form – probably smushed together by sound change and frequent usage – of Common, i.e. English or whatever language you are actually using at the table. This leads to slightly cliché but effective names like Greenwood, Highport, Bywater, Icewind Dale, and what have you. Some names might be the result of periods of occupation or settlement by speakers of other languages, however, and then it might be apt to either use some (phonologically adjusted, perhaps) version of those names, or else a translation of those names into the ‘Common’. But either way, those names need some kind of explanation why they are not simply in common: even foreign town names, when frequently used, tend to get adjusted to the language, so that English speakers say “Vienna” instead of “Wien” and “Moscow” instead of “Moskva”. Places within the English speaking region tend to have English names except in cases of settlement (like Native American adjusted terms in the US) or continuing occupation by speakers of another language, like dual names in Wales.

So when your map consists of a random jumble of names, with Riverdale directly next to Ta’Kannal and that in turn next to the Zaonar Mountains, you have made an unforced error. It probably won’t bother most people, but it’s worth considering how much more consistent and internally real your world becomes when there is some reason why the people of Riverdale don’t just call the Zaonar Mountains by a name of their own, say the Snowpeaks; is this because they are latecomers to the valley, and they adopted their interpretation of the name the local orcish clans gave the region (who may well not pronounce it quite Zaonar to begin with)? And how about Ta’Kannal? Besides the horrid fantasy apostrophe, why is this a name in a different style than the other? It points to yet another language, perhaps the name of some ancient figure who once owned or claimed the land but has since long been forgotten, except for the name that endured. (This certainly happened a lot with historic European placenames.) Immediately one’s imagination can start churning, and this in turn provides ample opportunity for deepening the world by introducing more difference and more consistency at the same time. Consider also the power that is in naming things: giving ‘official names’ to places that long had local names already was one of the sure means by which colonizers and explorers exercised dominance over conquered areas.


Religion. Another worldbuilding disaster provided by the default D&D approach is the notion of religion. I’ve personally never much liked the way divine magic is assumed to work in D&D anyway. Officially, clerics need not be clerics of gods, but can also gain their powers from ideas or philosophies. At the same time, all the actual spells and abilities are written in such a fashion that essentially the assumption of a one-on-one bond with a deity is there anyway, so that this clause does not really have much practical meaning. (One can imagine a priest-hating cleric of atheism in the default D&D world, perhaps, but what would be the holy symbol of a ‘cleric’ of utilitarianism?) It is not clearly explained how the spellcasting ability is supposed to relate to the deity itself, either, beyond a kind of quid pro quo arrangement. But neither of these things are a serious issue, and some of it is best left as unexplained mechanics – just as one does not really want to know how beholders evolve or how dragons retain combustible glands in their stomachs.

What is much worse is the ‘pantheon’ approach to the gods, as well as their meaning within the world in default D&D. The Dungeon Master’s Guide goes to great lengths to convince you that the key to religion in the default D&D world is to come up with some abstract domains covered by various gods and then to assign these to members of a pantheon, whom everyone then knows to be responsible for that domain and who (presumably) grant the powers associated with them. That this is a hackneyed and poorly considered pastiche of Graeco-Roman and Norse religion is one thing. Already here a problem arises: it is worth keeping in mind that for these ancient pagan peoples their gods were very real and part of an elaborate practice of religion, not of belief. Where the D&D notion assumes a kind of 19th century Protestant idea of religion as an individual relationship of trust in a deity as a one-on-one relationship, ancient religion was primarily a communal and community-building practice, centered around ritual sacrifice and the festival calendar, as well as the important offices of priesthood that usually doubled as significant sociopolitical positions.

The entire anthropological literature writing about the significance of mythology disappears out of the window in default D&D, since in their approach there is only really room for a mythical prehistory of divine intervention (akin to the Mesopotamian flood legends of the Old Testament) and very little for any ritual, community, or norm-sustaining stories, let alone narrative forms of natural philosophy, associated with deities, spirits, and natural powers as one actually observes in the world. This is a tremendous narrowing of the range of possibility of what religion might be. The same goes for the exclusion of the forms of religion that are not along the lines of the European model, such as the importance of spirit worlds inhering in nature, as in so-called animism, or the globally common and ancient practices of ancestor worship. No doubt inventive DMs have made use of these from time to time, but the default D&D setting does absolutely nothing to encourage you to even remember these exist, let alone point to ways to integrate them with the mechanics. Again, this is doing Tolkien one over, since he at least had the good sense – conscious of his own Christian beliefs and their importance to his very ability to do worldbuilding at all – to leave explicit reference to all these matters, or explaining their absence, out of most of his stories. (Admittedly not uniformly, and towards the end of his life he increasingly violated this rule, but that’s a different topic.)

That is not all, however. The most egregious principle of religion in default D&D, and to my mind one absolutely catastrophic for any kind of realistic and deeper worldbuilding in this respect, is the notion that these pseudo-pantheon deities actually exist, and that everyone knows it. This follows almost automatically from the way the divine magic system is set up by default, and very little is done to counteract it. Indeed, in most of the sample lore – although that is far from binding by itself – it appears to be evident that the gods are just facts of life and that everyone knows more or less who they are, what ‘domains’ they cover (as if they were a divine system of cabinet secretaries and ministers with or without portfolio!), and what things they have or haven’t done.

What this accomplishes is an imaginative disaster at several levels. Firstly, it robs the religious dimension of all of its remoteness and mystery, the element of distance to the divine and its unknowability – an element that is absolutely essential to its power and sublime mystique in the real world, regardless of what form of religion or spiritual practice is involved. Secondly, it eliminates another range of religious possibilities. One can’t very well be an atheist in such a world, except out of sheer stubborn cynicism, and neither can one be a monotheist, which is surely weird. But thirdly, and most importantly, it removes definitively the element of religious conflict in almost every way.

While the possibility still exists for conflict between adherents of this or that god, especially since it seems to be strongly suggested you produce both ‘good’ and ‘evil’ gods, very little else remains. The diversity of views about the content of religious belief is one of the most powerful dynamics of human history, for good or ill. Think of all the wars fought, the societies merging and splitting, the ink spilled, the cries to the heavens over whether a god or gods truly exist, and if so which one(s), and what they want from us, and how we should properly obey or worship them, and whether they interact with us directly or not, and who can represent them on Earth, and so on and so forth. All this is a fundamental dynamic of real human life. And in return for no gain whatsoever in depth, flavor, or even ease of gameplay, all of this is abandoned by default D&D in order to treat the gods as if they are like really big dragons, but wearing a beard and sitting on a cloud. It is as if the only possible imagination of religion is that taken directly from the illustrations in a not particularly clever Children’s Bible.

Again, I do not doubt that many a DM has realized that this approach has clear weaknesses, but going by what I observe on D&D streams and discussions of D&D settings on Reddit and elsewhere, the default D&D notion of religion is generally followed without much ado in the worldbuilding of most D&D games. To me, this is incomprehensible: if we are all agreed that the core aspect of narrative and storytelling is to set up conflict and to resolve it, in whatever fashion, why abandon one of the main dynamics of tension and conflict in actual human history? Lord knows fantasy economics is often silly and simplified enough, but that at least has a legitimate gameplay reason. And then everyone still recognizes that different folks have different attitudes to the importance of money, and things have different costs in different places, and haggling is a possibility. Default D&D religion robs us even of the equivalents of these basic premises, in exchange for… having gods with stats, or at least having your Lightning God give your cleric a storm spell. Why?


Economics. Speaking of economics, I do want to say a few things about it here. My criticism on this front is much more muted than the above two, which are perennial annoyances to me even in worlds that I otherwise enjoy a good deal (like Critical Role). When it comes to the economics of a D&D world, or perhaps better put its political economy in the classic sense, there are good reasons to simplify and what makes sense or not varies much more with what kind of world one intends. Obviously, not everyone wants to be particularly realistic in this regard: resource management is by no means fun to every kind of player, nobody wants to replicate the wage labor experience in one’s free time, and DMs have better things to do than compose tables of grain prices and trade nodes. (I should add that since I have a PhD in Economic History, I do not intend to knock tables of grain prices and trade nodes; those can be very informative!)

Even so, there are many small things that could be done to make things more realistic without adding great cost in time or complexity, and for larger things, it is worth at least thinking about them when you consider worldbuilding for your campaign. The small things can be something as simple as tossing out the arbitrary and absurd coinage system of default D&D, and realizing that nobody in medieval society ever used gold pieces as a standard currency – simply replacing gp with some invented silver coinage everywhere adds a great deal to realism without making it more difficult to keep track.

Depending on the type of setting, it may be worthwhile to think about whether you want a more (early to high) medieval approach, in which case you should not overpopulate your world with towns, or more of a high magic/tech late medieval/Renaissance approach, in which case one might expect higher agricultural productivity and therefore more people. But certainly consider that realistic medieval settings are poor and they are sparse, almost unimaginably so for us 21st century Western urbanites. In my view, the only fantasy setting that really replicates this feeling decently is The Witcher 3, and that is still compressed in space for the sake of gameplay. So think about where the towns are placed, how expensive transport overland is, where the food actually comes from, and finally, how magic actually interacts with the economy and production of your world. You don’t have to go overboard with this if you don’t want to, but it’s better to make conscious decisions than to just plop some towns on a map.

There’s a lot more that can be said about this, but those really require an elaborate post of their own, with a lot of conditionals depending on what type of high fantasy setting you want to play in. So on to the last topic of this rant.


Evil races. It is no secret, and I make it no secret, that I love Tolkien and his works. I think he was one of the most brilliant writers of all time, and he certainly more or less single-handedly created the genre of fantasy as we know it, even if certain people wrote imaginative stories before him. In fact, I think Tolkien’s work was such a remarkable accomplishment that we have trouble recognizing it. This is precisely because we are so used now to a situation in which ‘fantasy’ as a genre, and its tropes, are familiar to us and we can’t imagine what it would be like to hear the word ‘elf’ and not immediately have a set of associations about what that means (pointy ears, haughty, tall, woods, arrows, yadda yadda). It takes a lot of time and work to fully understand the context and the motivations that went into the making of Tolkien’s Legendarium, of which LotR and The Hobbit and even the published Silmarillion are only a part. I don’t think it can be done without studying the entire published History of Middle-Earth (in 12 volumes). But all that said, Tolkien did certain things for certain reasons, and while we can argue about how good those were, we certainly need not reproduce all of them. One of those is the idea of the ‘evil race’.

Default D&D has, as I mentioned, stepped away from many of its worst traditions in terms of the ‘race’ concept. It has removed the penalties for certain races and the class restrictions. In terms of race as used politically in societies in our real world, 5th Edition has certainly much improved the representation in terms of artwork. It has also published for the first time an explicit commitment to diversity and encourages players to experiment with and reflect on gender, sexuality, race, and such topics in their roleplaying. Given all this, I find it incomprehensible that both the main books and supplements such as Volo’s Guide to Monsters still maintain the notion of entire races of humanoid creatures that are considered to be, for all intents and purposes, inherently ‘evil’. Even more incomprehensible is how this notion is taken up, because part of default D&D, by many games in many D&D worlds, without further ado.

The sometimes kind of silly notion of what evil is in D&D aside (a topic for another day), it is clear that orcs, goblins, kobolds, hobgoblins and so forth are all essentially endowed with consciousness and understanding of the world identical to that of humanoids, whether or not they look more lizardlike or scaley or whatnot, and that they are in this respect no different from elves or dwarves conceptually. This makes it especially bizarre that the default lore, which especially in the Volo book is presented irritatingly definitively or ‘canonically’, pursues the hopeless task of trying to convince us that these beings are all (or 99%) entirely evil, craven, and selfish. It then adds insult to injury by trying to justify this with some poorly thought out and pulpy lore about how they are in the thrall of evil thief gods and the like.

Whatever we may think of them, Tolkien had reasons for his orcs being what they were, and even then his own attribution of evil to his creation bothered him his entire life. He fled into the murky depths of theological speculation to find a way to justify it and never quite satisfied himself. Much less then is there any reason for us, or for default D&D, to go even further on this path. It doesn’t even add anything indispensable to the game. Not only is it more narratively interesting if certain ‘races’ are perhaps feared or disliked but not actually inherently evil, indeed inherently anything; but for having bad guys who are just plain nasty, there are tons of aberrations, undead, and senseless monsters available, who are not humanoid and require no contorted justifications to depict as damned in some biblical sense. It is also easy enough to create bad guys from any range of humanoid races, goblins to humans, as individuals, without having to make weird Edwardian proclamations about the Nature of the Race. I can really see no good reason why this particular trope endures in this way, and why it is seemingly so hard to throw it overboard. I personally would find it a lot more enjoyable if we abandoned it, and let it go the way of the bikini armor babes of 1980s fantasy art.


The point of all this is not to say ‘your fun is wrong’. Rather, it is to use the opportunity of 5th Edition and the revival of D&D that accompanies it to encourage people to think about how their fantasy worlds work. High fantasy will always be based on tropes, and never be ‘realistic’ in any complete sense – that would defeat the purpose for sure. But that is no reason to endlessly repeat the same silly mistakes, or to simply copy bad practices and conventions because they have become the default of D&D, without thinking about it. Your fun is not wrong, but your wrong is not fun, or not as fun as it could be, I suspect: depth can be gained without anything being lost in exchange. It’s also an invitation to D&D writers and designers, DMs and others to think of ways to go beyond these cliches without having to make the game more complex or burdensome – especially those interested in that deep and cohesive world feel that attracts so many of us to fantasy literature and roleplaying in the first place. I have only scratched the surface here and pointed to some obvious and old mars and spots. There is no doubt a lot of work that can be done further exploring how we can integrate a better sense of world dynamics, even high fantasy world dynamics, into the framework of D&D that so many love.


The language thing is particularly bizarre given the way setting books work as an artform. Even in arty indie RPGs like Veins of The Earth and Maze of The Blue Medusa, which otherwise don’t miss a trick when it comes to shoving eyeball kick details into every table and statblock, they end up completely redundant. (And they never give you language families, though they have obvious gameplay ramifications: I’d almost forgive the silliness of having Goblin and Elf languages if they turned out to be mutually intelligible dialects of Fairy whose speakers pretend not to understand each other, a la BCMS)

The deity thing though – you played Baldur’s Gate right? I adore the Time of Troubles as portrayed in that game, as a kind of massively multiplayer version of The Bacchae where the gods wander the earth for a decade before pissing off back to Mount Olympus to be all abstract and omniscient, so that everyone is forced into a kind of dual consciousness where Cyrus is at once the Lord of The Underworld, The Final Judge of All Sinners and that jumped up prick from Highgate who slept with your wife back in ’46. It’s both a wonderfully fertile premise for a setting and an inadvertent but amusing pastiche of the whole Roman/Egyptian/etc practice of deifying former leaders. Annoying that the actual books never did much with it.

Have you read much Glorantha incidentally? Only setting I know that doesn’t reduce religion to a sort of pseudo-Bhakti mystery cult pyramid scheme nonsense.

Leave a Comment

Your email address will not be published. Required fields are marked *