Is thermodynamics a description rather than a mechanism?

kmarinas86

It seems to me that it should be. For example, some systems do not have a well-defined temperature. How do you use laws of physics depending on temperature (such as the law of thermodynamics) when you have a system that doesn't have a well-defined temperature? How do you even go about discussing things like thermodynamic irreversibility when your systems do not have a well-defined temperature?

K^2

You'd be surprised in how many systems you can define a temperature equivalent and apply thermodynamics. Traffic jams can be viewed as a phase transition, for example.

But you are right. It's a mathematical tool. More generally, what you are looking at is Statistical Mechanics, which gives you means of describing system consisting of very large number of particles or sub-systems that interact with each other. Even if interactions are simple, accounting for each one is impossible, but you can still talk about behavior of the system as a whole.

K^2

You'd be surprised in how many systems you can define a temperature equivalent and apply thermodynamics. Traffic jams can be viewed as a phase transition, for example.

But you are right. It's a mathematical tool. More generally, what you are looking at is Statistical Mechanics, which gives you means of describing system consisting of very large number of particles or sub-systems that interact with each other. Even if interactions are simple, accounting for each one is impossible, but you can still talk about behavior of the system as a whole.