Both the Scazzi case and the Kercher case would appear to illustrate the two fundamental current problems with the Italian criminal system:
Firstly, Italian criminal justice is still hamstrung by the remnants of the Fascist governments of the 1930s. The criminal codes were essentially rewritten under Mussolini's command to reflect his fascist views on law and order. Unfortunately for Italy, when democracy was restored after WW2, the existing criminal code was not just torn up (as it should have been), and replaced root-and-branch with a democratic code based on earlier pre-Fascist statutes. Instead, it was kept largely intact - with only the most egregious Fascist elements removed. However, the code still retained large elements of Mussolini-era statutes, most of which were terribly anti-libertarian, and redolent of a totalitarian state.
As a result, the criminal codes have taken decades to evolve slowly away from the Mussolini-era model. And the transition still has not completely taken place. For example, pre-trial detention periods are still far, far longer than nearly every other country - a clear throwback to Mussolini's fondness for codifying the practice of detention without trial.
The second problem with the current Italian criminal justice model is that it is an uneasy hybrid of the inquisitorial and adversarial systems. Inquisitorial justice is - in the view of most experts in jurisprudence - one of those concepts, like communism, that may seem utopian and ideal in theory, but which are unworkable and counter-productive in practice. It's naive and simplistic to think that the various parties in an inquisitorial justice system can be properly motivated by a simple desire to establish "the truth". Human beings live and think (and usually seek reward) in an adversarial way: as a a result, the best way to establish to establish accountability for a crime is to charge suspects and place them in front of an adversarial court system. If one party (the prosecutors) is arguing for guilt, and another party (the defence lawyers) is arguing for acquittal, the court has the best chance in practice of determining something close to "the truth". The adversarial system is not perfect, but it's demonstrably far, far better at achieving equitable, just results than the inquisitorial model.
The problem for Italy is that old concepts die hard. Even though the pure inquisitorial system is clearly riddled with flaws, the Italian legislature and judiciary appear to have been unwilling to dispense with it altogether. Instead, they've evolved an uneasy hybrid of the inquisitorial and adversarial approaches, which solves some problems but creates others. Even today, there appear to be many members of the Italian judiciary who find it hard to adapt to the "innocent until proven guilty" maxim of the adversarial approach - a maxim that is now finally enshrined in Italian law. And the court process still retains many unnecessary and somewhat-unworkable vestiges of the inquisitorial system - most notably the active representation of the victim (or victim's family) in the trial process, and the unhealthily close nature (from an adversarial perspective) of the relationship between judges and prosecutors.
It's my view - and, I'd suspect the view of many observers with an interest in criminal justice systems - that these problems have manifested themselves in the trial processes of both Scazzi and Knox/Sollecito. That's not to say that there aren't a number of good things about the Italian criminal justice system - for example the judges' post-verdict reports and the appeals system are both things that could benefit the anglo-saxon model (albeit at considerable potential expense). But I wonder if the ultimate acquittals of Knox and Sollecito in particular will shine a light onto what I would consider the two systemic problems with the Italian model.