# Why doesn't water get 'increasingly thicker' as it gets colder?

It's my understanding that the colder liquids get (or anything else for that matter) the slower the constituting particles move. That being the case, why is H$_2$O either 'water' or 'ice'? Given that temperature is continuous, why isn't there a continual physical change depending on temperature?

• Why should there be a "continuous change"? Different phases are mostly defined by such discontinuous changes - look up what a phase transition is. – ACuriousMind Aug 8 '15 at 12:52
• @ACuriousMind That's a textbook example of begging the question. – DanielSank Aug 10 '15 at 3:00

However, what's interesting about this curve is that it does not diverge as $T\to 0^\circ{\rm C}$. The reason is that a phase change really is for all effective purposes a discontinuous phenomenon: at $0+\epsilon^\circ{\rm C}$ the water molecules have enough kinetic energy to avoid binding to one another in a solid lattice: at $0-\epsilon^\circ{\rm C}$ they do not and this argument holds for any $\epsilon>0$. Otherwise put: the freezing point is an energy threshold in the same way that the behavior of a body in the neighborhood of the Earth shows behaviors that are similarly discontinuous functions of the body's total energy: if the body's speed energy exceeds the escape velocity at a point by an arbitrarily small amount, the body will follow a hyperbolic path and leave the neighborhood forever: lower the speed to an arbitrarily small amount below escape velocity and the path will be elliptic and the motion periodic: the path topologies change discontinuously from noncompact to compact.