The retroactivity of case law. Research on the fight against legal insecurity in private law

As retroactivity constitutes an application process of the law in time, it is worth first asking whether a judge's rulings are considered as a rule of law whether such case law is retroactive or declarative. To answer this question, it seemed necessary to define the rule of law as the rule that is intended to be used by a judge to settle a dispute. Then, to check whether the judge do make such rules of law, the Hart recognition rules were applied, inviting us to look at the attitude of the legislator, the judge and of the people to see if case law was considered as a source of law. The answer is yes with respect to the French Supreme Court's case law (Cour de cassation).
It then becomes necessary to question the cause of this retroactivity. The naturalistic theory, which provides  that any rule of law is naturally retroactive; and the mechanistic theory which justifies retroactivity by the need for the judge to apply the rule created by its ruling  to the dispute brought before him had to be excluded. The basis of retroactivity would be the incorporation theory, the application of which to case law as well as to changes in interpretation would be justified by the prohibition of regulatory judgements (arrêts de règlement).
Therefore one may wonder how to avoid the legal uncertainty produced by the retroactivity of case law ? Two solutions seem to be effective: either to enable the French Supreme Court to make regulatory judgements, or to introduce a kind of legislative summary proceedings enabling the French Supreme Court to request from the legislator to amend the rule, instead of creating a retroactive overruling decision. As it seems appropriate to maintain the complementarity between statutory law and case law, the creation of a legislative summary proceeding appears to be the only satisfying solution to the case law retroactivity issue.