Meeting a3-b4 plan with ...c4 in e5-chain structure - plan for black?

In some openings with an e5-chain, white will play a3, hoping to go b4. Black will sometimes meet this with ...c4

Here is an example from opening theory in the Advance French:

What is black's general plan here? I've played through games in this position, and have found some interesting ideas - sometimes black tries to occupy the b3 hole favorably, etc. I've seen games where he plays ...Ne7 and ...Nc8 to make room for kingside development, as well.

But in the long term, in such a locked position, doesn't black have to play ...b5, ...a5, and ...b4 to gain real play? And if so, isn't the queen terribly misplaced in b6? That was my understanding of this position, but most of the games I looked at didn't really feature this sort of slow build-up on the queenside.

Here's another example from a Caro-Kann game I recently played from the black side.

In short, I think I'm misunderstanding this idea and this sort of position, and playing over games or looking at it with Houdini doesn't help much, so I'm wondering if anyone has some insight as to how to play these positions!

These are a bit confusing to me too.

First of all, I wouldn't trust the computer's eval of c4. It makes the position very strategic and maneuvering so the engine has no idea, all it's seeing is space. This move changes the character of the position, one tench of a pawn is meaningless in the face of whether or not you like the ideas in the position.

Also, sure, the queen doesn't like b6, but now the pawn breaks for both sides are so difficult a tempo here or there makes no difference.

I looked at some games. In different games, both sides were breaking on either side. Sometimes white was playing b3 and sometimes black was breaking with b4. Sometimes white was going for f5 and sometimes black was expanding on the kingside.

I remember a comment from Soltis' book that when the board is divided where one side will have 3 files of play (the kingside here) vs 2 files, that that side with 2 files (the Q-side here) doesn't offer as much. That seemed to play out in the games as when white broke with b3 it didn't seem to mean much and it seemed black preferred kingside expansion.

The traditional breaks (b4 for black and f5 for white) took forever when they happened. I think one complication is that both sides castle opposite, and to the side where they had space. So the traditional breaks carried more drawbacks than usual. So my impression is this means there's no clear cut "I have to play this break even if it kills me" type of strategy. It reminded me more of an endgame actually, where each side carefully tries to secure a better piece here and there, or better control of a square. Action happened in slow motion. Even after lines were opened the better side was the one with better pieces (instead of the one who achieved their break first).

Anyway, I don't know much about it, but it's interesting. Hopefully someone more knowledgeable can offer a few pithy remarks for us.

I really like leiph's comment, and my comment might overlap his a bit.

Basically, yeah I don't think that just because ...c4 is a "queenside move" that therefore it must be followed by queenside play. It just kind of gains space. It prevents Bd3 for white for example. It also discourages white from playing b4, kind of fixing the pawn on b2, which can be awkward and have a cramping effect on white. Well, expanding on the queenside is certainly one way to go for black, but he can still play in the center with say ...f6, or push pawns on the kingside, if he thinks that would help him. There's no law against it. Maybe the b3 square will come in handy, maybe not, but ok, it's there.

Perhaps black just likes a closed position here and a little more space than he would otherwise have. ...c4 prevents white from opening the position with ideas like dxc5 followed by b4. Often one would say that since black no longer pressures white's d4 pawn, therefore white can build up on the kingside with a free hand. The problem with this is, again, black is not precluding himself from playing moves like ...f6, ...f5, ...g5, or even ...h5 to undermine white's pawns or something (e.g., in response to a g2-g4 push). Black doesn't even have to castle kingside. In fact, black's king might be safe on the queenside since it's hard for white to channel his pieces there or open lines there, due to the space black gained on the queenside. Black should just do what gives him a grip on the position, regardless of what side of the board it is on. I have looked at this french position for black before although not extensively. A lot of what black (and white for that matter) does just depends. Keep an open mind basically regarding the plans :)

"Also, sure, the queen doesn't like b6, but now the pawn breaks for both sides are so difficult a tempo here or there makes no difference."

Right. When the position is slow paced, we are thinking in the long term here. If, hypothetically, one side has a better overall strategy for example, losing a few tempi to carry out that strategy is not a fundamental problem.

In this case, when black played ...Qb6, he didn't necessarily know that white would try for a3 and b4. If white didn't play a3, maybe black wouldn't feel a need to play ...c4, but he couldn't know for sure what white would do when black played ...Qb6. Besides, the pressure on b2 does make it harder for white to move his c1 bishop :)

Yes, but is white's break (of f5) really that hard to achieve? It doesn't seem nearly as hard to achieve as black's b4, does it? The game in the second diagram I posted (my game) was sort of an example of this - it seems like white can play f5 long before black can make b4 effective, and f5 is also more dangerous. If black castles long, sometimes f7 is awkward to defend - though I guess just putting a rook on f8 is possible

As for playing on the kingside as black with ...f6 or something, I feel like it would require some sort of carelessness on white's part for this to actually be effective - wouldn't the most glaring feature of the position then be the backwards e6 pawn?

A lot of times black does indeed go for ...f6, even early on sometimes -- in fact only now do I recall black sometimes playing moves like  ...f6, ...Qc7 etc, sometimes ...g5-g4 to try to undermine e5, although my french memory is still a bit fuzzy. It makes weaknesses of course -- the e6 pawn and e5 square if white takes on f6, but it can also relieve the cramp for black. If you imagine a queenside castled black king, pushing some of the kingside pawns, maybe put a bishop on d6 (if white removes his pawn from e5), black could easily take over the initiative.

Black could also maybe play ...f5 and try to block things up that way. Some plans might be better than others, depending on the situation. Basically I'm saying that there are a lot of possibilities for black, particularly in the center and kingside, that shouldn't be discounted right away. And if black does go queenside I'm not sure I'm worried too much about g2-g4, f4-f5 ideas from white -- it could easily just give black a target.

One thing I was thinking about your game was that f5 should have been much harder. White seems like he really should be deficient on the light squares. I was thinking 16...Rb8 or 16...g6. I didn't understand you putting the knight on the b file (Nc8-b6).

...c4 is more of a blockading move than a thrust to play on the queenside. Black's king often castles long and Black's play is actually on the kingside.