of THEOPHRASTUS, and the master of PTOLEMY PHILADELPHUS. As little, and as ill as we are informed of the ftate of natural philosophy among the more antient naturalifts of Greece, fuch asPYTHAGORAS, Anaxagoras,DEMOCRITUS, and others, whofe names are preferved tho' their works are loft, we know enough of PLATO and ARISTOTLE, whose works have been preserved, perhaps more to the detriment than to the advancement of learning, to determine what the state of it was in the days of STRATO. We know that it was no longer the study of nature by obfervation, and experiment; but that it confifted in a jargon of words, or at best in some vague hypothetical reasonings: and yet STRATO, who could not have told the egyptian king how the idea of purple, the color of his robe, was produced, pretended to account for all the phænomena, and, among other doctrines, to establish that of the plenum, for he laughed at the vacuum, as well as at the whole atomical system of DEMOCRITUS.

HYPOTHESES are much in the favor of fome philofophers; for there have been many STRATOS even among the moderns. But hypotheses may be employed without being abused. In all our attempts to account for the phænomena of nature there will be fomething hypothetical neceffarily included. The analytic method itself, our fureft road to science,does not conduct us further than extreme probability, as it has been observed; and this probability must stand us in lieu of cer

tainty.

tainty. But when we cannot arrive by this method at such a probability, is it reasonable to make an hypothefis? Is it reasonable, when we cannot draw from obfervation and experiment fuch conclusions, as may be safe foundations on which to proceed by the fynthetic method in the pursuit of truth, to affume certain principles, as if they were founded in the analytic method, which have been never proved, nor perhaps fuggested by the phanomena, in hopes that they may be so afterwards? In a word, when the only clue we have fails us, which is most reasonable, to stop short, or to push forwards without any clue at all into the labyrinth of nature? I make no fcruple of deciding in a case, so plain, that it would be a filly affectation of modesty, not modefty, to hesitate, When the phænomena do not point out to us any fufficient reason why, and how a thing is as we discover it to be, nor the efficient cause of it, there is a sufficient reafon for ftopping fhort, and confeffing our ignorance; but none for feeking, out of the phænomena, this reason, and this cause which we cannot find in them. This is learned ignorance, of which the greatest philofophers have no reason to be afhamed. "Rationem-harum gravitatis proprietatum ex phænomenis nondum potui "deducere, et hypotheses non fingo," said our NEWTON, after having advanced natural knowledge far beyond his cotemporaries, on the fure foundations of experiment, and geometry. How preferable is this learned ignorance to that ignorant learning, of which fo many others have foolishly

а

foolished boasted? DES CARTES, who mingled fo much hypothetical with fo much real knowledge, boasted in a letter to his intimate friend the minime MERSENNE," that he should think "he knew nothing in natural philosophy, if he "was only able to fay how things may be, with"out demonstrating that they cannot be other"wife." LEIBNITZ, who dealt in little else than hypotheses, speaking, in his reply to BAYLE'S Reflections on his Pre-established harmony, of the ridiculous whimfy of his monades, and the rest of his metaphyfical trash, compares himself to ANTAEUS; afferts that " every objection gives "him new ftrength;" and boafts vainly that he might fay without vanity,

Omnia præcepi, atque animo mecum ante peregi.

It will be urged, perhaps, as decifive in favor of hypotheses, that they may be of fervice, and can be of no differvice to us, in our pursuit of knowledge. An hypothefis founded on mere arbitrary affumptions will be a true hypothesis, and therefore of service to philosophy, if it is confirmed by many obfervations afterwards, and if no one phænomenon stand in oppofition to it. An hypothesis that appears inconfiftent with the phænomena will be foon demonftrated false, and as foon rejected. This reafoning, which is the fum of all that can be faid for them, will not hold good, I think, in either case, enough to countenance the abuse of them which is made by the very persons who urge this plea in favor of them,

That

That such an hypothesis may be true, is within the bounds of poffibility; because it implies no contradiction to fuppofe that men, who pass their lives in gueffing, may guess fometimes right. A man may throw ten fixes with ten dice; but no man in his senses would lay that he did, nor venture his stake on such a chance. In the other cafe, it is true that an hypothefis inconfiftent with the phænomena may be foon demonftrated falfe. But it is not true that it will be as foon rejected. If philofophers are fond of making hypotheses, their disciples are as zealous to defend them. The honor of a whole fect is thought to be engaged, and every individual is piqued that another should fhew that to be falfe, which he has all his life taken to be true; so that, notwithstanding all the graces of novelty, a new truth will have much to do to diflodge an old error. Instances of this fort are innumerable. Let us produce one from astronomy itself.

IF any hypothesis was ever affumed with a plaufible probability, that which we call the ptolemaic was fo. The apparent face of the heavens led men to it. We may say that the phænomena fuggefted it, and that the revolution of the fun, planets, and stars, in feveral spheres round the earth, could scarce be doubted of by men who affumed any general conclufions, instead of drawing them all from a long courfe of particular obfervations carefully and learnedly made. The plaufibility of this falfe hypothefis, and the autho

rity of the peripatetic school, established it on the ruins of the true fyftem which PYTHAGORAS had brought long before into Italy from the east, and which was probably that of the egyptian, and chaldean aftronomers. Falfe as it was, it maintained it's credit thirteen or fourteen centuries, if we reckon only from the time of the alexandrian aftronomer PTOLEMY to that of COPERNICUS. Many difficulties had occurred, but as fast as they did so new affumptions were made to reconcile them, till the whole became one complicated heap of hypothefis upon hypothefis. It was banished at last, and a truer system took it's place. The fautors of hypotheses would have us believe that even the detection of their falfhood gives occafion to our improvement in knowledge. But the road to truth does not lie through the precincts of error, and the improvement of aftronomy was not owing to the deftruction of the ptolemaic hypothefis; but the destruction of this hypothesis was owing to the improvement of aftronomy. If this hypothefis had never been made, COPERNICUS Would not have had the honor of reviving the pythagorean fyftem, but mankind would have had the benefit of pursuing, without interruption, a system founded on knowledge, inftead of pursuing, during an interval of so many centuries, an hypothesis founded on affumption.

To this antient let us join a modern instance, to suggest the same reflections, and confirm the fame proofs. The fyftem of DES CARTES dazzled

and