Humanoid hybrids in D&D 5e

I would like to review the evidence for interbreeding between different humanoid races in official core 5e DnD material. I suggest that the amount of interbreeding implies that several of these races qualify as a single species, if such a scientific term can be applied to fantasy creatures. (This point is relevant to a few other things I may post in future, and to points I sometimes want to make in online discussions, and I’m posting it mainly so I can refer back to it as a sort of footnote.)

(Later note: this turned out to be an early post in a series on race topics. I added a summary post signposting them all afterwards.)

Humans, Orcs and Ogres

Human-orc hybrids are described in the PHB (Player’s Handbook) race chapter as half-orcs. Human-ogre hybrids are described under Ogre in the Monster Manual (MM) as half-ogres. And ogre-orc hybrids are described under Ogre (and referred to under Orc) in the MM as half-ogres (specifically ogrillons). So it is clear that there is a mutually-interbreeding core of humans, orcs and ogres.

Other interfertile humanoids: human-kin

Returning to the PHB, human-elf hybrids are described in the race chapter as half-elves, and it is noted in the Human section (under the heading A Broad Spectrum and alongside notes on the great variability within the human race) that ‘A lot of humans have a dash of nonhuman blood, revealing hints of elf, orc or other lineages.’ This line implies both that humans hybridize with at least one and probably multiple additional races, and that part-human hybrids are themselves fertile, in order to breed again with humans and produce descendants with mainly but not entirely human ancestry. It also suggests that these trace ancestries produce visible differences in their descendants, becoming part of the general range of human variability.

The Orc entry in the MM says that orcs ‘readily crossbreed with other races’ and goes on to specify that they cross-breed with dwarves and other ‘humanoids of similar size and stature’ to orcs, producing orcs or half-orcs.

Ogre-hobgoblin and ogre-bugbear hybrids are described under Ogre in the MM as half-ogres. The line ‘Ogres don’t mate with dwarves, halflings or elves. They eat them.’ in the half-ogre paragraph could be read as implying that ogres biologically could mate and hybridize with these races if they chose.

So at least one of the core three can specifically interbreed with each of elves, dwarves, hobgoblins and bugbears, and possibly halflings. My understanding of real-world cross-breeding is that if A can breed with B and B can breed with C, then A and C can (usually) breed with each other so, between that and the facts that ogres have four listed and three implied potential breeding partners, and that humans and orcs at least can interbreed with an unspecified range of additional humanoids, I am inclined to infer that all the races mentioned in bold in this paragraph can probably, in principle, breed with each other. I’m going to call these interbreeding races collectively ‘human-kin’ for ease of reference now.

Speculative additions

Are there other races we can add to the human-kin?

Goblins are stated in their MM entry to form the ‘goblinoid’ family with bugbears and hobgoblins, and so seem likely to be potentially capable of interbreeding with whatever those other goblinoids can breed with. That is specifically ogres, so perhaps there’s just too great a size difference for goblins to breed directly with them. But it would seem plausible for goblins to be fertile with goblinoid half-ogres, and with orcs, humans and whatever other human-kin are of a compatible size with them.

Drow are mentioned under Elf in both the PHB and MM, so I am inclined to treat them as elves for interfertility purposes. ‘Aquatic elves’ are mentioned in the sahuagin MM entry, and I guess would be another sub-race. (The sahuagin themselves might very speculatively be transformed aquatic elves because of their malenti variant, but for now I will more conservatively discuss them under humaniform animals towards the end of this post.)

Duergar are described in the MM as descended from dwarves, though not listed under them. Likewise grimlocks from humans, and perhaps ettins from orcs. It would be plausible (though I don’t think it is strictly demanded) to treat these also as interfertile with the same partners as their respective ancestral races.

Halflings, apart from the one doubtful line under half-ogre, are not stated to interbreed with any other race. But their only major physical difference with humans is size, and I feel it is reasonable to include them as human-kin if desired, subject to size differentials.

Gnomes are also not positively stated to breed with any other race. But there is no evidence that they are sufficiently physically different from halflings and dwarves that you would expect them to be less interfertile. Svirfneblin could be treated as a variety of gnome for this purpose.

Giants are typically humanoid in their physical characteristics, apart from their size. Indeed, ogres have the giant type and so this type is clearly not a barrier to interfertility, and if ettins are descended from orcs then populations can move between the humanoid and giant types. (Given the stated antiquity of giants in the MM, world-builders could even decide that smaller humanoids are the diminished descendants of original giants.) I therefore don’t think it would be too much of a stretch to treat at least some giants as human-kin, subject to size differences.

Gith have humanoid type and physique, features broadly within the range of the above humanoid races (if you use psionics), and an origin story rather like duergar or grimlocks, which I think leaves it open to include them in a wide human-kin heritage pool, if that is what you want to have.

Nonhumanoid heritage

There are strains of humanoid that are stated to have nonhumanoid ancestry, or that have elements of nonhumanoid creatures in their description. There is little evidence that many of them are interfertile with other humanoids, and I am inclined to treat it as a default under the core rules as written that they do not interbreed with human-kin.

The one exception, with positive evidence for nonhumanoid interfertility with human-kin, is fey (PHB, elf race entry, fey ancestry feature). I think it is left open whether elves were already a distinct variety of humanoid when they acquired their fey ancestry, or whether they were ordinary humans and now owe most of their distinctive features to fey ancestry. Either way, it is open for world-builders to decide whether (some, most or all) fey can naturally interbreed with humanoids. If not, the ‘fey ancestry’ could be conceived as a magical influence rather than natural breeding.

Elves are outnumbered by the variety of humanoids with features of a nonhumanoid creature and a magical origin story, and no positive core rulebook evidence for the potential to interbreed naturally with other races (beyond being ‘humanoids of similar size and stature’ to orcs). Pending evidence to the contrary, I am therefore inclined to take it as a default position that these humanoid types may not themselves breed with human-kin. Examples include dragonborn, tieflings and yuan-ti. Some yuan-ti are not even of the humanoid type, and all are stated in the MM to have ‘utterly sacrificed their humanity’.

Some other creatures with humanoid features, such as driders, lamias and medusas, are stated to be individually produced by magical transformations with no positive evidence that they can reproduce at all. Minotaurs in contrast can reproduce (explicitly with each other, implying that they are not fertile with non-minotaurs). Harpies have a similar magical origin, but it is not clear whether they are individually cursed like driders or are self-reproducing like minotaurs (I think the latter is a more obvious reading of the MM) . I see no canon evidence that any of the creatures in this paragraph can back-cross with humanoids, though it isn’t ruled out either

Half-dragons, according to the MM, are born of pairings between humanoids and dragons in humanoid form. They (alone among core 5e hybrids as far as I see) are explicitly infertile.

Cambions are stated by the MM to be born naturally of matings between humanoids and fiends in compatible shape. They are not stated to be infertile; if they were interfertile with humanoids cambion ancestry could provide an alternative origin story for tieflings (though this would somewhat contradict the tiefling PHB entry).

Other creatures that are made up of human-shaped parts and animal-shaped parts (centaurs, merfolk and merrows) are not clearly explained in their MM entries (merrows are altered merfolk, but there the trail stops). However, there is no positive evidence of interfertility with complete humanoids.

Lack of humanoid heritage

There are some creatures that have humanoid shape and type, but are stated to be transformed versions of nonhumanoids and so are implied to have no human ancestry or common biological heritage. These include gnolls and jackalweres.

In the absence of evidence to the contrary, I think it would be reasonable to extend this to other anthropomorphs such as aarakocra, bullywugs, kenku, kobolds, kuo-toa, lizardfolk, and thri-kreen. (Also I guess quaggoths, sahuagin and troglodytes though their animal base seems less specifically identifiable.) Even less likely to breed with natural humanoids are examples (such as ettercaps) that don’t have the humanoid type.

Species status

Although I like to use common-sense realism in fantasy (like fire burning and water flowing downhill, unless author-chosen fantastical elements intervene), I don’t usually like to apply modern scientific concepts to fantasy worlds. Science is sufficiently complex that attempts to mix it with fantasy will lead to unwanted inconsistencies, often pointed out by someone who knows more science than the author. But some people like to apply the concept of ‘species’ to fantasy races, so I’ll just touch on that here.

A common definition of species in modern science is a population of individuals that can potentially produce fertile offspring together. This applies, in core 5e, to human-kin as discussed above: humans, orcs, ogres and others. The proper term for the named, distinguishable types within this species would then be sub-species or race (and race is indeed what the rulebooks call many of them).

There are, to be fair, other definitions of species. I’m not going to discuss those that rely on genes and chromosomes, as I really don’t think that this level of science fits in DnD. Those based purely on morphology (the form and appearance of creatures) are I think rather dated conceptually, and used by modern scientists mainly when they can only examine remains and cannot analyze genes or reproduction. But there are variants on the reproductive definition that look at whether the creatures in question choose each other as mates or recognize each other as potential mates.

By this mate-choice definition, ogres are canonically not the same species as dwarves, elves or halflings, because the MM states that they do not mate with them.

This definition does introduce a complication in that ogres do mate with humans, and humans do mate with elves. And ogres do mate with orcs and orcs do mate with dwarves. To apply the mate-choice definition or for any other reason to rule out selected hybridization pairings (if that is what you want to do), we need to construct a more complex model. I feel that a workable scheme for this case would be hub-and-spoke. Humans, orcs and ogres are the hub. Elves are a spoke attached to humans but not ogres. Dwarves are a spoke attached to orcs but not ogres. Hobgoblins and bugbears attach to ogres. The other human-like races such as halflings, gnomes, goblins etc. may or may not attach as spokes somewhere on the hub. Some at least ought to fit with orcs and probably humans to comply with the general notes that these hub races interbreed with a range of other humanoids as well as those specifically mentioned. It is also a question whether any of the spoke races interbreed with each other; personally I would say that goblins, hobgoblins and bugbears should at least.


No GM is tied to the rules or lore as written; you can choose to have more or less hybridization in your world.

More hybridization could start by making the most generous interpretation of all the areas of ambiguity in the canonical material: all human-kin can interbreed with all others (rather than only the specifically-mentioned hybrids being viable); halflings are indeed interfertile with human-kin (subject perhaps to size); drow, duergar, grimlocks, ettins and so on are equivalent to their main race cousins; goblinoids are treated for hybridization purposes as one race with three sizes; and the default is that humans and orcs (and therefore ogres) can indeed hybridize with most other humanoids of the right size.

Then you can add hybridization potential where none is mentioned in the core books, but it would be plausible based on physical similarity: I think gnomes (and svirfneblin) would be the first to add, then giants (starting with the most basic like hill giants and continuing as far through the giant type as you wish) and perhaps gith.

Then, led by the fey ancestry of elves, you could include in the human-kin bloodline various strains with non-humanoid origin. These influences could include fiendish, celestial, and draconic. It would require a few tweaks to canon, but I think it would make sense to say that direct matings or initial magical crosses produce half-human-kin (cambions, half-dragons, etc.) and that recombining these hybrids with original human-kin produces varieties such as dragonborn and tieflings then, with further dilution, human-kin with a dash of exotic ancestry. Likewise you could mine the monster manual for other nonhumanoid heritages to blend into human-kin by repeated hybridization, inspired by half-animal creatures like yuan-ti, driders, harpies, merfolk and centaurs, and animal-like creatures with humanoid shapes like kenku, kobolds or gnolls.

On the other hand of course you could have less hybridization. First you could reject all hybrids not explicitly mentioned in canon, so there would be no human-dwarf or orc-elf crosses. This would involve disregarding the directions in the human and orc descriptions they they hybridize with unspecified others, or at least explaining them away as negligible. If you want to finesse some agreement with canon and the general principle of transitivity in interfertility, you could say that while biologically possible, they never happen because those races do not choose to mate with each other.

You could go further against canon by removing from your setting hybrids like half-ogres or orc-dwarves mentioned in the MM, or even finally by dis-including the well-known PHB hybrids: half-elves and half-orcs. (Edit: I’ve now argued in another post that something like this decision may be kind of necessary if you want these races to remain separate over many generations of contact with each other.)


What are your thoughts on this issue? How do you handle hybridization in your world? Have I missed any relevant data in the core 5e books? (I know there are points in other 5e books that I don’t own, and that things worked differently in older editions and also do in other games and stories, and these could also be interesting points of discussion.) Let me know in the comments section.

Leave a comment

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