Recently I saw a post on TikTok about how there's no good ways to "tank" in D&D 5th edition; that the role isn't supported by the system. This is an interesting statement because it touches on several elements of game mastering and game design that are important:
Forced action is rarely fun in roleplaying games.
To maximize fun, the Game Master should "buy in" to player choices.
Positive challenges are more reliable than negative challenges.
Let's dive in to each of these topics using "tanking" as our lens of analysis. But first, a quick bit of terminology for those who aren't familiar (or might agree there is no such thing as tanking in 5e): my use of the terms "tank" or "tanking" are focused on a playstyle that is concerned with helping the party by mitigating damage as it happens. This playstyle doesn't rely on healing other people who take damage, but typically revolves around two objectives: attract the attacks of enemies to yourself; survive the damage dealt by those attacks.
Those familiar with MMORPGs (a.k.a. MMOs) and other video game formats are probably familiar with the idea of "tanking", along with the related concept of a "taunt": any ability that accomplishes the first objective of tanking, which is to attract enemies to hit you instead of other characters / targets. Taunts in MMOs are typically "hard taunts", i.e. they force an enemy to attack the tank and give no options for counter-play. More about that later.
Forced Action is Unfun
A primary concern in tabletop RPGs is making sure players have autonomy - that they have the ability to choose what their characters do. This could be as simple as what they order at the tavern or as complex as how they plan to depose a corrupt ruler. When players aren't given enough choice - enough control over their character and the story's progression - that is called "railroading". In other words the party is moving along the predetermined tracks of a railroad that the GM has designed, with little or no ability to influence where the train (story) goes next.
Something that doesn't get talked about often enough is that the GM also needs to have autonomy to have fun in a roleplaying game. The GM is a player too, even if they have more responsibilities and narrative power than the other players. I say that to directly respond to a common complaint about tanking in D&D 5th edition, which is that there are no taunts. When people say this, I think they are speaking specifically about hard taunts. And while they are correct in a literal sense that hard tuants don't really exist, that is because it would be bad for them to exist. The GM should be allowed to make decisions appropriate to the narrative and not feel compelled to take any specific action regardless of story. This is why spells that deprive choice, like dominate person and other enchantment effects, require saving throws. The character gets a chance (or multiple chances) to resist that effect and maintain their ability to choose.
So yes, there aren't a plethora of hard taunts in D&D 5e. But that's because their shouldn't be; they go against the core principles of RPGs and inhibit fun.
GM Buy In
You might have heard this sentiment expressed by the phrase "shoot the monk", which comes from the notion that monks have an ability that allows them to negate or even reflect ranged attacks. This ability often results in GMs deciding to never shoot the monk since it makes bad things happen to the monsters they control. But, as the phrase suggests, this is a mistake. Player abilities are supposed to be cool, impactful, and often do bad things to the GM's monsters. That is by design. So by avoiding the ability, the GM is effectively removing that ability from the game and depriving the player of a mechanic that could have been part of why they wanted to play that character.
Similar logic is held up to contest the notion of tanking in D&D. "If the tank is designed to mitigate attacks and make them less effective, then why would I attack them?" Because the tank can accomplish their task of surviving lots of attacks, either by causing them to miss or by having enough hit points to not care about the damage caused by said attacks, the GM is disincentivized to attack them. Combine that with the lack of hard taunts in the game and we arrive at the conclusion that tanks can't exist.
But that is the trap of failing to buy in to player abilities. When a player expresses interest in playing a character that is designed to tank, the GM should instead welcome that decision and make it a priority to let the player tank. I.e. they should shoot the tank! The player is clearly interested in tanking and is deriving some level of fun from doing so. By ignoring that decision and failing to value their desire to tank, the GM is directly causing a reduction in fun.
This is why D&D has plenty of "soft taunts" - abilities that give the GM / monsters good reasons to want to focus the tank without entirely negating their autonomy. Most features utilize some form of granting disadvantage on attacks if the attack doesn't target the tank. This is great because it gives the enemy a tactical advantage to keep hitting the tank. Which is better: dealing some damage to a character that is designed to take damage; or dealing no damage to anyone? Sure it's possible to still hit with disadvantage (and thus the GM still has the ability to choose), but it stacks the deck against the monster and gives them a good excuse to continue buying in to the tank's source of fun.
Positive vs Negative Challenge
So let's say the GM does buy in. They attack the tank, and in so doing combats become too easy. Or perhaps they simply become unfun for the GM, whose monsters must hopelessly and pitifully strike at an unkillable barbarian or too-heavily-armored fighter. Didn't we just say that GMs are players too and should be able to have fun?
This is where the other part of "buying in" comes in. When a GM wants to challenge their players, they can do so in a positive way or in a negative way.
Negative challenges are monster abilities or circumstances that negate the features or choices of players. An example would be creating an enemy whose strikes deal force damage instead of bludgeoning damage, thus bypassing the barbarian's rage resistances. The GM is buying in to the barbarian's desire to be the target of attacks, but is punishing that decision by negating the thing that makes the barbarian good at taking attacks.
Positive challenges, on the other hand, are abilities or circumstances that challenge a player not by negating or preventing their powers from working, but by meeting those powers with unique abilities of their own. Creating challenges that are so powerful that the players need to use everything at their disposal to achieve victory. For example, instead of making monster attack bonuses so large that they negate the heavy armor a fighter has invested in, use a spellcaster to cast heat metal on the fighter. Part of their build identity has been acknowledged (wanting to be the target and having high AC), but other parts have been challenged (being able to ignore that AC to deliver damage). The fighter can still run around and utilize their armor to deflect blows from other enemies, but the heat metal spell will add a ticking clock of constant damage that makes the fight more intense.
Both positive and negative challenges can be deployed effectively, and neither is inherently good or bad. From what I've seen negative challenges are typically easier to implement and often fail to give players as much buy in as they would want, meaning they can become unfun if used too much. Positive challenges are more difficult to design and implement, but can help give parties the epic hero moments they need to feel awesome. A good game uses both.