Reflexions made on the foregoing Paper by Mr. John Greaves , Savilian Professor of Astronomy in the University of Oxford .
1645 .
THis Reformation of the Roman Calendar , Proposed by Mr. Dee , as I cannot wholly approve , so I cannot altogether disapprove .
For I like the Subtraction of Ten Days , as the Church of Rome has done , beginning the Computation from the Council of Nice : though it cannot be denied , but that the Reformation from the time of our Saviour had been much better But since the Fathers of the Council of Nice thought it more Wisdom to look forwards , than to look backwards , and to have greater Care of avoiding Distractions in the Church , about the Celebration of Easter for the future , than to remedy the Errors past : I think we shall do well , with the Church of Rome to follow their Example .
And whereas some have thought of a more exact Calculation , than this Emendation , introduced by Pope Gregory the xiiith .
which they ground upon the late Astronomical Observations of the learned Tycho Brahe : yet since the Difference is not so great , as to make any sensible Error in many Ages , and since that Error may be easily corrected by the Omission of an intercalary Day , I think it not fit for so small a nicety to make a new Dissension in the Church .
Much less am I of their Opinion , who think this Correction of the Year therefore to be rejected , because it comes recommended by the Church of Rome : which were all one to refuse some wholesome Potion , because it is prescribed by a Physician whose Manners we approve not of .
And thus far I assent to Mr. Dee .
But I cannot subscribe to his Opinion , that this Reformation should be made by the subtraction of ten days out of one year alone .
For tho ' I grant , that this were a quick cure of a lingring Disease , yet it is against all Rules of Art in curing one malady to make Ten .
For it cannot be , but that the Defalcation of Ten Days in one Year must be of infinite Disturbance in the Common wealth in all Contracts , where necessarily a certain time is defined .
And therefore when Julius Caesar the Dictator corrected the Roman Year by the help of Sosigines , a Mathematician , aster this manner , that is , by Subtraction of Days , that Year , in which he did it , was called by the Ancients Annus Confusionis : by Reason of the great Confusions and Inconveniences , which thereby happened : and I doubt not , but that the Year 1582 .
in which the Defalcation of Ten Days was made by the Bull or Edict of Pope Gregory , might justly also be styled Annus Confusionis .
But such Examples , as these , are not to be imitated .
For what Caesar did as Dictator , or what Gregory the xiiith .
did as Pope , the one by the Power of the Temporal Sword , the other of the Spiritual Sword , is not to be practised by Gracious Princes .
I shall therefore humbly recommend to His Majesty 's Wisdom , and favourable Consideration , that Course , which was long since proposed by many able Mathematicians to Pope Gregory , upon the first Notice of his Purpose of Correcting the Calendar ; which if it had been known , either to Mr. Dee , or to his Learned Judges , or to the Wise and Honourable Lord Burleigh , the Reformation with us had long since been finished , and not one Man prejudiced in his Estate .
The manner was this ; that for Forty Years space there should be no Bissextile or intercalary Years , or as we call them Leapyears , inserted in the Calendar .
By which course it is most evident , that ten Days will be Subtracted in forty Years , and these forty Years will be each of them anni aequabiles , consisting of 365 Days , as our common and ordinary Years do , without any alteration in the whole Year .
And this being beyond all Exception , had been readily entertained by Pope Gregory , had not his Ambition been greater than his Judgment ; for he was willing to have the Honour of this Emendation , and not to leave it to his Successors ; whereby the Year ever since has been called Annus Gregorianus .
My Opinion therefore is , that by His Majesty 's Letters Patents , some Skilful Astronomer should be appointed to have the Compiling and Publishing , within His Majesty 's Dominions , of all Calendars and Almanacks for forty Years , in which space , by omitting the Intercalations , we shall at length come to agree with the account of the Church of Rome : and every Year , during this time of Forty Years , shall be as this present Year 1645 .
and as those of 1646 .
and 1647 .
will be in the usual and ordinary computation .
III .
A Calculation of the Credibility of Human Testimony .
MOral Certitude Absolute , is that in which the Mind of Man entirely acquiesces , requiring no further Assurance : As if one in whom I absolutely confide , shall bring me word of 1200 l accruing to me byGift , or aShipsArrival ; and for which therefore I would not give the least valuable Consideration to be Ensur 'd .
Moral Certitude Incomplete , has its several Degrees to be estimated by the Proportion it bears to the Absolute .
As if one in whom I have that degree of Confidence , as that I would not give above One in Six to be ensur 'd of the Truth of what he says , shall inform me , as above , concerning 1200 l : I may then reckon that I have as good as the Absolute Certainty of a 1000 l , or five sixths of Absolute Certainty for the whole Sum .
The Credibility of any Reporter is to be rated ( 1 ) by his Integrity , or Fidelity ; and ( 2 ) by his Ability : and a double Ability is to be considered ; both that of Apprehending , what is delivered ; and also of Retaining it afterwards , till it be transmitted .
What follows concerning the Degrees of Credi``bility , is divided into Four Propositions .
The Two First , respect the Reporters of the Narrative ; as they either Transmit Successively , or Attest Concurrently : the Third , the Subject of it ; as it may consist of several Articles : and the Fourth , joins those three Considerations to``gether , exemplifying them in Oral and in Written Tradition .
PROPO S. I. Concerning the Credibility of a Report , made by Single Successive Reporters , who are equally Credible .
LET their Reports have , each of them , Five Sixths of Certainty ; and let the first Reporter give me a Certainty of a 1000 l , in 1200 l : it is plain that the Second Reporter , who delivers that Report , will give me the Certainty but of 5/ 6ths , of that 1000 l or the 5/ 6th of 5/ 6ths of the full Certainty for the whole 1200 l. And so a Third Reporter , who has it from the second , will transmit to me but 5/ 6ths of that Degree of Certainty , the Second would have delivered me &amp;c. That is , if , a , be put for the Share of Assurance a single Reporter gives me ; and , c , for that which is wanting to make that Assurance complete ; and I therefore supposed to have a/ a +c of Certainty from the First Reporter ; I shall have from the Second , aa/ ~a+c 2 ; from the Third , a 3/ ~a +c 3 .
And accordingly if , a , be = 100 ; and c= 6 , ( the number of Pounds that an 100l , put out to Interest brings at the Years end ; ) and consequently my Share of Certainty from One Reporter , be = 100/ 106 ; which is the present value of any Sum to be paid a Year hence : The Proportion of Certainty coming to me from a Second , will be 100/ 106 multiplied by 100/ 106 , ( which is the present Value of Money to be paid after two Years ; )and that from a Third-hand : Reporter , = 100/ 106 , thrice multiplied into itself ; ( the Value of Money payable at the end of Three Years , ) &amp;c. Corollary .
And therefore , as at the Rate of 6 per , Cent Interest the present Value of any Sum payable after Twelve Years , is but half the Sum : So if the Probability or Proportion of Certitude transmitted by each Reporter , be 100/ 106 ; the Proportion of Certainty after Twelve such Transmissions , will be but as a hall ; and it will grohw by tat Time an equal Lay , whether the Report be true or no. In the same Manner , if the Proportion of Certainty be set at ~ it will come to a half from the 70th Hand : And if at 1000/ 1001 , from the 695th .
PROPO S. II .
Concerning Concurrent Testifications .
IF Two Concurrent Reporters have , each of them , as 5/ 6ths of Certainty ; they will both give me an Assurance of 35/ 36ths , or of 35 to one : If Three ; an Assurance of 215/ 216 , or of 215 to one .
For if one of them gives a Certainty for 1200 l , as of 5/ 6ths ; there remains but an Assurance of 1/ 6th , or of 200 l wanting to me , for the whole .
And towards that the Second Attester contributes , according to his Proportion of Credibility : That is to ~ of Certainty before had , he adds 5/ 6ths of the 1/ 6th which was wanting : So that there is now wanting but 1/ 6th of a 1/ 6th that is 1/ 36th ; and consequently I have , from them both , 35/ 36ths of Certainty .
So from Three , 215/ 216 , &amp;c. That is , if the First Witness gives me a/ a +c of Certainty , and there is wanting of it c/ a + c ; the Second Attestes will add a/ a +c of that c/ a + c ; and consequently leave nothing wanting but c/ a +c of that a/ a + c , ~ c2/ ~a +c 2 .
And in like manner the third Attester adds his a/ a +c of that c2/ ~a + c| 2 , and leaves wanting only c 3/ ~a +c 3 .
&amp;c. Corollary .
Hence it follows , that if a single Witness should be only so far Credible , as to give me the Half of a full Certainty ; a Second of the same Credibility , would ( joined with the first ) give me 3/ 4ths ; a Third , 7/ 8ths ; &amp; c : So that the Coattestation of a Tenth , would give me 1023/ 1024ths of Certainty ; and the Coattestation of a Twentieth , 2096999/ 2097000ths or above Two Millions to one .
&amp;c. PROPOS .
III .
Concerning the Credit of a Reporter for a Particular Article of that Narrative , for the whole of which he is Credible in a certain Degree .
LET there be Six Particulars of a Narrative equally remarkable : If he to whom the Report is given , has 5/ 6ths of Certainty for the whole , or Sum , of them ; he has 35 to one , against the Failure in any One certain Particular .
For he has Five to One , there will be no Failure at all : And if there be ; he has yet another Five to One , that it falls not upon that single Particular of the Six .
That is , he has ~ths of Certainty for the whole : and of the ~th wanting he has likewise 5/ 6ths , or 5/ 36ths of the whole more ; and therefore that there will be no Failure in that single Particular , he has 5/ 6ths and 5/ 36ths of Certainty , or 35/ 36ths of it .
In General , if a/ a+c be the Proportion of Certainty for the whole ; and m/ m+n be the chance of the rest of the particular Articles m , against some one , or more of them n ; there will be nothing wanting to an absolute Certitude , against the not failing in Article , or Articles , n , but only nc/ ~m+n~a+c PROPOS .
IV .
Concerning the Truth of either Oral or Written Tradition , ( in Whole , or in Part , ) Successively transmitted , and also Coattested by several Successions of Transmittents .
( 1 ) SUpposing the Transmission of an Oral and Narrative to be so performed by a Succession of Single Men , or joined in Companies , as that each Transmission , after the Narrative has been kept for Twenty Years , impairs the Credit of it a th part ; and that consequently at the Twelfth Hand , or at the end of 240 Years , its Certainty is reduced to a Half ; and there grows then an even Lay ( by the Corollary of the second Proposition ) against the Truth of the Relation : Yet if we further suppose , that the same Relation is Coattested by Nine other several Successions , transmitting alike each of them ; the Credibility of it when they are all found to agree , will ( by the Corollary of the first Proposition ) be as 1023/ 1024 of Certainty , or above a Thousand to One ; and if we suppose a Coattestation of Nineteen , the Credibility of it will be , as above Two Millions to One .
( 2 ) In Oral Tradition as a Single Man is subject to much Casuality , fo a Company of Men cannot be so easily supposed to join ; and therefore the Credibility of 100/ 106ths , or about 19/ 20ths , may possibly be judged too high a Degree for an Oral Conveyance , to the Distance of wenty Years .
But in Written Tradition .
the Chances against the Truth or Conservation of a single Writing are far less ; and several Copies may also be easily supposed to concur ; and those since the Invention of Printing exactly the same : several also distinct Successions of such Copies may be as well supposed , taken by different Hands , and , preserved in different Places or Languages .
And therefore if Oral Tradition by any one Man or Company of Men might be supposed to be Credible , after Twenty Years , at 19/ 20ths of Certainty ; or but 9/ 10ths , or 4/ 6th : a Written Tradition may be well imagined to continue , by the Joint Copies that may be taken of it for one Place , ( like the several Copies of the same Impression ) during the space of a 100 , if not 200 Years ; and to be then Credible at 100/ 101ths of Certainty , or at the Proportion of a Hundred to One .
And then , seeing that the Successive Transmissions of this 100/ 101 of certainty , will not diminish it to a H , al until it passes the Sixty ninth Hand ; ( for it will be near Seventy Years , before the Rebate of Money , at that Interest , will sink it to half : ) It is plain , that written Tradition , if preserved but by a single Succession of Copies , will not lose half of its full Certainty , until Seventy times a Hundred ( if not two Hundred ) Years are past ; that is , Seven Thousand , if not Fourteen Thousand Years ; and further , that , if it be likewise preserved by Concurrent Successions of such Copies , its Credibility at that Distance may be even increased , and grow far more certain from the several agreeing Deliveries at the end of Seventy Successions , than it would be at the very first from either of the Single Hands , ( 3 ) Lastly in stating the Proportions of Credibility for any Part or Parts of a Copy , it may be observed ; that in an Original not very long , good Odds may be laid by a careful Hand , that the Copy shall not have so much as a Literal Fault : But in one of greater Length , that there may be greater Odds against any Material Error , and such as shall alter the Sense ; greater yet , that the Sense shall not be altered in any Considerable Point ; and still greater , if there be many of these Points , that the Error lights not upon such a single Article ; as in the

