Aristotle canvasses these tables at some length, and amends them partly by making the fourth case of the second table change place with the fourth of the first.[41] He then discusses whether we can correctly say that the _necessary to be_ is also _possible to be_. If not, then we might say correctly that the _necessary to be_ is _not possible to be_; for one side or other of a legitimate _Antiphasis_ may always be truly affirmed. Yet this would be absurd: accordingly we must admit that the _necessary to be_ is also _possible to be_. Here, however, we fall seemingly into a different absurdity; for the _possible to be_ is also _possible not to be_; and how can we allow that what is _necessary to be_ is at the same time _possible not to be_? To escape from such absurdities on both sides, we must distinguish two modes of the Possible: one, in which the affirmative and negative are alike possible; the other in which the affirmative alone is possible, because it is always and constantly realized. If a man is actually walking, we know that it is possible for him to walk; and even when he is not walking, we say the same, because we believe that he may walk if he chooses. He is not always walking; and in his case, as in all other intermittent realities, the affirmative and the negative are alike possible. But this is not true in the case of necessary, constant, and sempiternal realities. With them there is no alternative possibility, but only the possibility of their doing or continuing to do. The celestial bodies revolve, sempiternally and necessarily; it is therefore possible for them to revolve; but there is no alternative possibility; it is not possible for them not to revolve. Perpetual reality thus includes the unilateral, but not the bilateral, possibility.[42] [Footnote 39: Aristot. De Interpret. p. 21, a. 34-p. 22, a. 13. See the note of Waitz, ad Organ. I. p. 359, who points out the error of Aristotle, partly indicated by Ammonius in the Scholia. The rule does not hold in propositions with the sign of universality attached to the subject; but it is at least the same for Modals and Non-modals.] [Footnote 40: Aristot. De Interpr. p. 22, a. 14-b. 28.] [Footnote 41: Aristot. De Interpr. p. 22, b. 22, [Greek: lei/petai toi/nun] &c.; Ammonius, Schol. p. 133, b. 5-27-36. Aristotle also intimates (p. 23, a. 18) that it would be better to reverse the order of the propositions in the tables, and to place the Necessary before the Possible. M. Barthélemy St. Hilaire has inserted (in the note to his Translation, p. 197) tables with this reversed order.] [Footnote 42: Aristot. De Interpret. p. 22, b. 29-p. 23, a. 15.]