How exactly does "specific beats general" function?

[hacksnake]

Caveats: Long-time lurker; first time poster. I have a little background in advanced logic / set theory sort of stuff.

Trying to keep it short but it's long anyway... how do you determine which rule is more specific?

Example from PHB:
"...many adventurers don't have proficiency with longbows, but every wood elf does because of a racial trait."

So, I break this down into sets of things & find the most specific as such:

Set of all things.
Set of all things that happen to also be adventurers.
Set of all adventurers that happen to be wood elf.


In order from top to bottom each is a proper subset of the next one up.
You look at rules that apply to each in order and then apply them.
Therefore wood elf adventurers have proficiency.

"I'll still be an adventurer (i.e. - a PC) even if someone polymorphs me into a dwarf" etc.

Example from these boards I've found problematic:
BM Ranger riding companion. For this I broke the sets down as:

Set of all things.
Set of all things that are also beast, medium or less, & CR 1/4 or less (i.e. - valid BM companions).
Set of things that are actual BM companions.
Set of things that are actual BM companions being ridden at the moment.


The short story is: "It's always my companion but it's not always my mount. When it's a mount is more specific than when it's my companion."

From that understanding it's clear that a mounted companion follows mount rules where there's a disagreement between them & companion rules.

That said...
The discussion has obviously not been so cut & dried on this and other topics.
That makes me think maybe I'm misreading / misunderstanding how "specific beats general" works.

Mostly I'm interested in the general problem of "how do you determine what's more specific in a way that's not really debatable?" vs. the specific rules question about the mounted BM. That case just serves as a useful platform from which I can explain my general problem.

Thoughts? Is there any existing clarification about how specific beats general is supposed to be ruled?

Thanks!

[Mjolnirbear]

Pets not attacking on their own unless you spend a resource is a general rule. Familiars and BM companions both use this rule. So does mounted, unless the mount is independant. The rule might go : any permanent pet or mount you have cannot attack unless you spend a resource to let it.

Specific would be, for instance, Improved Familiars. They are improved forms granted only to warlocks and to casters that find these creatures in the wild and negotiates.

Specific would also be Find Steed. You summon them, they are permanent, and as intelligent creatures they are always independant. In this case it's because the spell says the steed is in complete accord with the paladin that rides them; they have a mental bond.

Animate Dead is not an exception. You cast a spell to make them and a spell to keep them. Summon spells as well: you spend a strong spell to get them and they have a limited duration. Neither are permanent pets.

Edit: my off-the-cuff possible rule has been edited to take into account the point made in the post below this one.

[Elbeyon]

Your familiar acts independently of you, but it always obeys your commands. In combat, it rolls its own initiative and acts on its own turn. A familiar can’t attack, but it can take other actions as normal.Familiars are completely independent and get a full set of actions. They just can't use that action to attack.

[Mjolnirbear]

The mount rule in the BM mount example allows the BM a way to allow his pet to attack. The independant mount rule is specific; the permanent pet rule in this case is general.

The PP rule seems to be an example of the action economy rule. Getting extra attacks for free is hard; it's givrn to martials specifically but otherwise you need to spend a bonus action, which you only have one of. Rangers already get the extra attack for free. Giving another free extra attack through the companion treads on fighter-only territory.

[Mjolnirbear]

Familiars are completely independent and get a full set of actions. They just can't use that action to attack.

Hmm. You're right. Id have to edit my PCP rule then. But the essence of my point remains; it's an example of a pet that you cannot attack with.

[hacksnake]

Thanks for the replies... I might have answered my own question typing out this response.
Rather than asking about the Attack action - let's look at Dash.

Consider a mastiff (Mike) and a halfling BM Ranger (Roger). Here's 4 scenarios to consider:

Roger purchased Mike & isn't riding him right now. Mike is not Roger's companion.
Roger purchased Mike & is riding him right now. Mike is not Roger's companion.
Roger purchased Mike & isn't riding him right now. Mike is Roger's companion.
Roger purchased Mike & is riding him right now. Mike is Roger's companion.


Given those scenarios and considering the Dash action RAW is, I believe:

No RAW supports Roger making Mike Dash that I saw.
Roger can make Mike Dash without spending his own action (Controlling a Mount - PHB 198).
Roger can make Mike Dash by spending his own action (Ranger's Companion - PHB 93).
Roger can make Mike Dash without spending his own action (Mike as a mount is a more specific scenario than Mike as a companion).


I've seen that 4th bullet debated even though it seems like clear RAW to me.
Some adventurers also happen to be wood elves - so they get proficiency with some weapons.
Some companions also happen to be mounts - so they get to take Dash without using the Ranger's action (but only while actually mounted).

So here's the core of it:
My criteria for applying "Specific Beats General" is to analyze the scenario and see which rule is associated to which subset & then the inner most subset is the rule that "wins" & then upwards in order. This tells me clearly for a given scenario which rule is more specific - by telling me which subset is more specific.

So my core question is: am I doing it wrong?

The only language about which rules are specific that I see is:
General rules:

...rules, especially in parts 2 and 3...

Specific rules:

...many racial traits, class features, spells, magic items, monster abilities, and other game elements break the general rules...

So someone might say "Mounted Combat is a general rule because it's in part 2; Ranger Companions are a specific rule because it's a class feature - so RAW you can't gain benefit by mounting your companion."

So is that it? Anything in chapter 2 & 3 is explicitly a "General Rule" & everything else is a "Specific Rule"?

That seems a little sketchy to me because of loose language like "especially" and "other game elements".

[Mjolnirbear]

It's general if it applies to multiple cases. So it's a matter of perception.

Mounted combat could be seen as a general rule. All players can take advantage of it, and any animal that can serve as a mount is subject to it. Ranger, in this case, could be the specific exception to the rule.

The Point of View that allows a ranger BM mount to attack says that Mounted is the specific rule; in this point of view, BM companion is a general rule that covers almost all permanent pets and also action economy.


Which one is general and which is specific? Who knows? In either case, it's the DM's interpretation that matters. Generally you want the rule that applies to everyone to be the general rule, and the rule that applies only in certain circumstances (such as a hiding halfling) to be specific. In the case where either rule could be specific or general, well, there is no heirarchy, no rule; there's just interpretation, which is why the original Mounted BM post suggested that it wouldn't work unless your DM specifically agreed with your interpretation.

[LordVonDerp]

One minor note, the elf keeps longbow proficiency after being polymorphed.

[hacksnake]

One minor note, the elf keeps longbow proficiency after being polymorphed.

I don't think that's accurate.

From polymorph:

The target's game statistics... are replaced by the statistics of the chosen beast...

Are racial traits not "game statistics"?

[hacksnake]

It's general if it applies to multiple cases. So it's a matter of perception.

Mounted combat could be seen as a general rule. ...

Which one is general and which is specific? Who knows? ... Generally you want the rule that applies to everyone to be the general rule, and the rule that applies only in certain circumstances (such as a hiding halfling) to be specific. In the case where either rule could be specific or general, well, there is no heirarchy, no rule; there's just interpretation, which is why the original Mounted BM post suggested that it wouldn't work unless your DM specifically agreed with your interpretation.

I see what you're saying.
I think we're approaching it from two perspectives.

One says that rules themselves are either general or specific:
From this stand point I think what you're saying makes a lot of sense that it's just interpretation.
You could say:
"Mounted" is a specific rule for "Creature".
"Companion" is a specific rule for "Creature".
So mounted companion is specific vs. specific - RAW tells you nothing.

Likewise you might say that:
"mounted is a general rule since anyone can do mounted stuff to a creature"
"companion is a specific rule since it's class feature"
"Therefore, RAW says you can't get the free Dash."

So it all comes down to how you decide which one happens to be a "specific rule" or a "general rule". If you decide "mounted" is a general rule then it's always general for purposes of SBG determination. Hopefully I got what you were saying correctly.

One says rules themselves are neither general nor specific except in the context of a situation:
From that perspective it's very clear that "mounted" is more specific than "companion" (except in contrived scenarios like below).
Beasty is your companion whether or not it's mounted & the mounted state keeps changing while the companion state remains static. So the companion rule applies to more cases than the mounted rules & is therefore clearly more general.

An exception would be some contrived scenario in which you are grafted to your mount and some magic can toggle the state of 'companion-ness'. In that case it's clear that mounted is more general. In this case you're mounted on it whether or not beasty is your companion & the companion state keeps changing.

Only in a scenario where "companion-ness" and mounted state both change round-to-round is it then up to interpretation because in actuality neither is more general than the other.

I think the conclusion I'm coming to is basically:
"RAW doesn't really say exactly how Specific Beats General functions - so it's up to each play group".

I don't see anything compelling RAW-wise that supports my interpretation of "SBG" as being more or less valid than the others expressed here.

So that means SBG is functionally useless to settle RAW disputes unless everyone involved happens to agree on which viewpoint to take regarding all the rules in question & in that case you probably don't have a rules dispute to begin with :).

Thanks again!

[LordVonDerp]

I don't think that's accurate.

From polymorph:


Are racial traits not "game statistics"?

I don't know, but if we start letting polymorph re-write a target's mind then problems are sure to arise.

Also, has it been clarified that you can still use you hands if you're polymorphed into another person? I remember there being some confusion about that awhile back.

[coredump]

Specific and General are *not* a characteristic of a particular rule. It is a relative relationship between two rules. It is like saying "tall beats short". Looking at Human and Kobold, Human is the 'tall' and wins. Looking at Human and Goliath, Goliath is 'tall' and wins.

There are situations that arise where there is no way to determine which is 'more specific'.....luckily those are pretty rare.

[Icewraith]

Usually "specific beats general" is when an ability lets you do something normally against the rules. For instance, only the fighter's extra attack class features stack with each other in contravention of the usual rule that extra attacks don't stack.

Similarly, someone with the Polearm Master feat can make opportunity attacks under more generous circumstances than the ones spelled out in the movement section. However, affecting that character with the shocking grasp spell denies them the ability to make reactions at all. Shocking grasp is more specific than Polearm Master which is more specific than the standard attack of opportunity rules.

[hacksnake]

I don't know, but if we start letting polymorph re-write a target's mind then problems are sure to arise.

Also, has it been clarified that you can still use you hands if you're polymorphed into another person? I remember there being some confusion about that awhile back.

Well, RAW-wise I think the spell does literally & explicitly rewrite your mind - "...including mental ability scores..." insofar as your INT, WIS, & CHA scores somehow represent "your mind". The only thing it explicitly states you keep is your alignment & personality.

It is probably a balance thing to prevent historical polymorph abuse by stacking retained mechanics onto a more powerful physical chassis. It also prevents things like a Druid from wild shaping out of a hostile polymorph (unless a Druid is a shapeshifter - in which case they automatically succeed the save to begin with).

Polymorph forms are limited to beasts so in my OP I should have said "...true polymorphs me into a dwarf...". I have no idea regarding rules clarification around using hands. In the same sentence RAW says you are limited by the new form and you explicitly can't take any action that requires hands or speech. Great apes have use of hands and various animals (such as parrots) are physically capable of speech.

Personally & without looking it up, I think I'd parse the RAW to be restrictive since it uses ", and" joining two independent clauses. From a "common/correct English usage" perspective it seems clear that you are limited by the new form and also you can't do this list of explicit things. Neither clause is clearly meant to supersede or clarify the other (as I read it).

To further support that as RAW (& maybe even RAI) the wording on True Polymorph explicitly adds the clause "...unless its new form is capable of such actions." to the bit about being unable to take certain actions.

I think it guts some fun out of polymorph and I also think that's how the rules are written & probably intentionally so given how broken 3.x polymorph could be.

EDIT: me do grammar good.

[hacksnake]

Specific and General are *not* a characteristic of a particular rule. It is a relative relationship between two rules...
That's how I initially read it although I added "...relative relationship between two rules in a specific context."

I was surprised to see some of the rules arguments in other threads that seemed to me easily solvable by applying specific-beats-general.

After more thought I took another look at the brief text in the PHB. I could read it either way if I put myself into a frame of mind that wanted it to read one way or another ("rules implicitly are always general or specific" or "it's all situational"). So that didn't make me feel great that the rule is useful :).


Usually "specific beats general" is when an ability lets you do something normally against the rules. For instance, only the fighter's extra attack class features stack with each other in contravention of the usual rule that extra attacks don't stack.

Similarly, someone with the Polearm Master feat can make opportunity attacks under more generous circumstances than the ones spelled out in the movement section. However, affecting that character with the shocking grasp spell denies them the ability to make reactions at all. Shocking grasp is more specific than Polearm Master which is more specific than the standard attack of opportunity rules.
I think that's consistent with how I'd rule the situation as well given the "build subsets & find deepest nested one" approach.
In this case you end up with two 'set chains':

Opportunity Attacks
Opportunity Attacks under the effects of Shocking Grasp



Opportunity Attacks
Opportunity Attacks caused by Polearm Master
Opportunity Attacks caused by Polearm Master under the effects of Shocking Grasp


The most specific scenario is "Opportunity Attacks caused by Polearm Master under the effects of Shocking Grasp". That set is the intersection between the other sets "Opportunity Attacks under the effects of Shocking Grasp" & "Opportunity Attacks caused by Polearm Master".

I think you are forced to view the ordering of the sets in this scenario in the 2nd grouping because that specific opportunity attack doesn't exist at all without polearm master feat. Once you've established that the OA can exist next you see that sometimes it's under the effects of shocking grasp and sometimes it isn't.

If that's how the rules work then it makes me feel good about my approach.

The other interpretations would, I think, need to say "it's up to your DM how Polearm Master & Shocking Grasp interact" beacause "OAs by PM" & "OAs under SG" are both "specific" & "OAs" are general. Also PM is a feat & SG is a spell - spells are specifically cited as an example of "specific"; feats are not although they may be an "other game element".

So that is at least some reasonable circumstantial evidence that applying some sort of situational analysis is a more correct approach.

EDIT: just adding - the thing I'm most interested in is how to properly determine which rules are "more specific" & ideally with some form of RAW support or overwhelming circumstantial evidence or something. If all RAW gives us is "it's all just a matter of opinion which is more specific so let's bicker about that instead" then the rule may as well not exist (my 2cp).

I like the responses - they are all helping me think about it from more angles so thanks! :)

[numerek]

BM Ranger riding companion. For this I broke the sets down as:

Set of all things.
Set of all things that are also beast, medium or less, & CR 1/4 or less (i.e. - valid BM companions).
Set of things that are actual BM companions.
Set of things that are actual BM companions being ridden at the moment.

Beast master ranger companions

Set of all things.
Set of all things that are also beast, medium or less, & CR 1/4 or less (i.e. - valid Beast master ranger companions).
Set of things that are actual Beast master ranger companions.

Mounts

Set of all things.
Set of all willing creatures that are larger size than their rider with appropriate anatomy.
Set of things that are currently mounted.

Beast master ranger companion mounts

Intersect of all sets above.


Neither list item 3 above is a proper subset of the other.

If we can also look at sets of actions.
Mount actions(and other actions)

Set of all actions.
Set of all actions players can perform.
Set of all actions players can perform if they have the things required to perform those actions.

Class actions

Set of all actions.
Set of all actions players can perform.
Set of all actions players can perform if they have the class ability.

Beast master companion actions

Intersect of all action sets above.


In the actions sets beast master with their companion actions is a proper subset of mount actions(and other actions) and therefore more specific that any player with a mountable creature.

[hacksnake]

Beast master ranger companions

Set of all things.
Set of all things that are also beast, medium or less, & CR 1/4 or less (i.e. - valid Beast master ranger companions).
Set of things that are actual Beast master ranger companions.

Mounts

Set of all things.
Set of all willing creatures that are larger size than their rider with appropriate anatomy.
Set of things that are currently mounted.

Beast master ranger companion mounts

Intersect of all sets above.


Neither list item 3 above is a proper subset of the other.

That's certainly true.
I think that's a really good representation of the "treat each chain independently" approach. What it misses is any sort of analysis about the existence of the specific creature.

So the way I'd build on this example is to consider the situation of the specific creature & through which chain they get into the various subsets, how often they fit into each, & etc. Which sets do they drift in and out of? Which sets are they permanently a member of? For most intents & purposes anyway even if not literally permanently a member (ex: polymorph removing elf-hood, something severing companion-state, etc).

In the companion example you see that over time they drift in and out of the 'set of things that are currently mounted' but they persist in the 'set of things that are actual Beast master ranger companions'. Even in the language you need to use to construct the sets it's clear that being mounted is more temporary than being a companion. I think it's reasonable to conclude that temporary statuses are "more specific" than longer lived ones (is it?).

For me that means that being a companion is clearly a "more general truth" of their existence than "being mounted" is. Caveats above about contrived scenarios which would lead me to judge differently.

I think it's the same if you consider the situation from the perspective of the ranger. It's more generally true that this beast is the companion than the mount.

So either way "it's my mount" is a more specific scenario. I tried to save it by adding "...and is this specific animal" as a lower set (which ends up being a proper subset of the other chain as well) but that approach goes both ways.

Good call on them not being proper subsets of each other. I think the approach is still salvageable through this lens, no? I need to expand beyond subset analysis & add in some more information viewing the sets over time to fish out which is more general - so I suppose the whole thing gets a bit more sketchy... it again drifts further into "well, now let's quibble about the details of the scenario & what they mean in the context of these set hierarchies".


If we can also look at sets of actions.
Mount actions(and other actions)

Set of all actions.
Set of all actions players can perform.
Set of all actions players can perform if they have the things required to perform those actions.

Class actions

Set of all actions.
Set of all actions players can perform.
Set of all actions players can perform if they have the class ability.

Beast master companion actions

Intersect of all action sets above.


In the actions sets beast master with their companion actions is a proper subset of mount actions(and other actions) and therefore more specific that any player with a mountable creature.

Clever.

So the whole "just throw things into sets" bit is clearly insufficient. Good call.
You can construct the sets from various perspectives to gain the outcome you want & then you can quibble about why this or that representation is right/wrong.

[numerek]

That's certainly true.
I think that's a really good representation of the "treat each chain independently" approach. What it misses is any sort of analysis about the existence of the specific creature.


I am not a set theory expert so sorry if I don't get the significance of the distinction but if you are looking at the specific creature it wouldn't really matter what sets its in because all of the sets that it is in would all have the same cardinality(1) They would all be equal sets from my view.


In the companion example you see that over time they drift in and out of the 'set of things that are currently mounted' but they persist in the 'set of things that are actual Beast master ranger companions'. Even in the language you need to use to construct the sets it's clear that being mounted is more temporary than being a companion. I think it's reasonable to conclude that temporary statuses are "more specific" than longer lived ones (is it?).


presumably the creatures starts its existence neither as a companion nor mounted.

As time goes on the creature can become mounted and can be subsequently dismounted.

As time goes on the creature can become a beast master ranger companion.

As time goes on the creature can become a mounted beast master ranger companion.

Besides by dying there is no RAW way for the creature to stop being a beast master ranger companion.

So the creature can be mounted regardless of being a beast master ranger companion and can be east master ranger regardless of being mounted. Again I see no distinction in specificity.



For me that means that being a companion is clearly a "more general truth" of their existence than "being mounted" is. Caveats above about contrived scenarios which would lead me to judge differently.

I think it's the same if you consider the situation from the perspective of the ranger. It's more generally true that this beast is the companion than the mount.

So either way "it's my mount" is a more specific scenario. I tried to save it by adding "...and is this specific animal" as a lower set (which ends up being a proper subset of the other chain as well) but that approach goes both ways.

Good call on them not being proper subsets of each other. I think the approach is still salvageable through this lens, no? I need to expand beyond subset analysis & add in some more information viewing the sets over time to fish out which is more general - so I suppose the whole thing gets a bit more sketchy... it again drifts further into "well, now let's quibble about the details of the scenario & what they mean in the context of these set hierarchies".

As far as permanency that is further complicated by choices if the beast is mounted before becoming a beast master ranger companion and never dismounted then it was mounted longer that it was a beast master ranger companion.



Clever.

So the whole "just throw things into sets" bit is clearly insufficient. Good call.
You can construct the sets from various perspectives to gain the outcome you want & then you can quibble about why this or that representation is right/wrong.

Perhaps I was being to clever and did not do a good job constructing my action sets as I did not take into that you must be mounted in order to do mount actions which was obviously taken into account in the other set analysis and why it is unable to determine specificity. and if you apply it to the action sets you get the same ambiguity.

[TheOOB]

I usually explain it using magic the gathering. In magic a creature cannot attack the turn it comes into play. That's a general rule. Some creatures have an ability called Haste which says the creature can attack the turn it comes into play, that's a rule specific to the card. When you summon the creature the two rules conflict, one rule says the creature cannot attack the turn it comes into play, one rule says it can. Since haste is a more specific rule, it wins and the creature can attack the turn it comes into play.

A D&D 5e example, the rule for maintaining concentration when damaged is to make a con saving throw with a dc of half the damage, minimum 10. Lets say I hit you with a spell that requires you to make a con save with a dc equal to the total damage delt. In that case the spells wording overrules the general wording in that case.

[hacksnake]

I am not a set theory expert so sorry if I don't get the significance of the distinction but if you are looking at the specific creature it wouldn't really matter what sets its in because all of the sets that it is in would all have the same cardinality(1) They would all be equal sets from my view.
I agree. I tried applying the specific creature to the sets themselves & it doesn't help exactly as you said.

What I was going on about in that bit was kind of a side-analysis that uses the sets as an input.


presumably the creatures starts its existence neither as a companion nor mounted.

As time goes on the creature can become mounted and can be subsequently dismounted.

As time goes on the creature can become a beast master ranger companion.

As time goes on the creature can become a mounted beast master ranger companion.

Besides by dying there is no RAW way for the creature to stop being a beast master ranger companion.

So the creature can be mounted regardless of being a beast master ranger companion and can be east master ranger regardless of being mounted. Again I see no distinction in specificity.

Right. The creature "can" all sorts of things. However the creature actually "did" only specific things:

the creatures starts its existence neither as a companion nor mounted.
As time goes on the creature may or may not have become mounted and subsequently dismounted.
As time goes on the creature actually became a beast master ranger companion.
As time goes on the creature sometimes becomes mounted and subsequently dismounted.


From that list the chain of events is more clear since we're talking "what actually happened here?" vs. "what could have happened to get us to this point?"


As far as permanency that is further complicated by choices if the beast is mounted before becoming a beast master ranger companion and never dismounted then it was mounted longer that it was a beast master ranger companion.

Agreed. In that case I'd be forced to rule that mounted is more general according to my proposed method. It's admittedly a silly conclusion so it may be evidence that this method isn't very good.



Perhaps I was being to clever and did not do a good job constructing my action sets as I did not take into that you must be mounted in order to do mount actions which was obviously taken into account in the other set analysis and why it is unable to determine specificity. and if you apply it to the action sets you get the same ambiguity.

I meant clever as a positive, "well done sir, I hadn't noticed this weakness in my approach before" :smallsmile:.


I usually explain it using magic the gathering...
I think the M:tG rules & mechanics are far more structured due to the tournament scene. The whole game is built so it's not ambiguous which is more general and which is more specific: rules in the rules are general & rules on the cards are specific. I think something like that is actually in the rules although it's been at least a year since I read through the M:tG rules. I think it's phrased more like "card text modifies/beats rules text".

As I recall there are also codified metarules dictating which wins in scenarios like "you can x" and "you can't x" etc. Further all the card text is written within this framework to attempt to avoid ambiguity.

If 5e had some metarule that said "Restrictive Beats Permissive" (or the other way around) then I think the companion question in particular would be a non-discussion. I think for 5e this approach becomes more difficult due to: the lack of other helpful metarules, and lack of clear guidance on what's a "specific rule" (i.e. - "no card text is specific; rules text is general" distinction).

From the earlier discussion I think if rules themselves are general or specific it's entirely fair to say "Mounted is specific. Companion is specific. RAW says nothing about this interaction."
You could shift perspective and it's just as accurate to say "Mounted is general. Companion is specific. RAW says you're bound to companion actions."

You could probably construct many other conclusions based on the RAW. Again, I think it's because we don't have a clear delineation between which rules are "specific" or "general". In M:tG the cards trump the rules & that's a rule in the rules (as I recall). In 5e "the cards" are part of "the rules" so it's all intermixed. So how do we figure out which is which?

EDIT: just to elaborate a bit about why this is a sticking point for me - I think the 5e rules did a terrible job of explaining what they meant here with any precision. Spells are called out as likely candidates for being both specific and general.

Parts 2 & 3 are given as examples of where general rules live - spells live in part 3.
Spells are called out as being specific rules later.

So I need to apply Specific Beats General to successfully parse the text of the Specific beats General rule itself? /table-flip :smallbiggrin:

EDIT2: I think it I M:tG-ified the companion scenario then 'the card' would list both "Companion - <to character>" & it'd also say "Mountable". The rules would go into detail about how those two things function mechanically. That rules text would either address the interaction or we'd fall back to metarules about "can" vs. "can't" & such.

In that case Companion 'wins' probably because it's arguably reasonable to say "Companion really reads something like you can't take any actions except for blahblah" & mounted says "you can take x actions".

I don't think that tells us much about how DND 5e works though but that's how I think it'd play out, roughly speaking, if this were M:tG.

[Giant2005]

General rules are rules that apply to everyone and everything. Specific rules are rules which apply only to a single race/class/feat/whatever.
With respect to the BM Companion and Mounted Combat rules, the Mounted Rules apply to everyone that feels like riding a mount while the Companion rules apply exclusively to the Beast Master. Due to the companion rules being specific, they take precedence over the Mounted combat rules but that doesn't mean the Mounted Combat rules don't apply, it only means that any Mounted combat rules that are mutually exclusive with the Companion rules don't apply.