XI .
A Letter to the Rev. Dr. William Brakenridge , Rector of St. Michael Bassishaw , London , and F. R. S. concerning the Term and Period of Human Life : In which the Inequalities in constructing , and the false Conclusions drawn from Dr. Halley 's Breslau Table are fully proved ; the supposed extraordinary Healthfulness of that Place is particularly examined , and confuted ; and its real State equalled by divers Places in England ; the Imperfection of all the Tables formed upon 1000 Lives is shown ; and a Method propoposed to obtain one much better : By T. W. A. M. It were to be wished , that some inland town could be found in England , where there was kept an annual register of births and burials , with the ages of the deceased , and where there is no confluence of strangers .
' ' Dr. Brakenridge , in Phil. Trans. Vol. XLIX .
p. 172 .
An author should be fond of reading his works to those , who know how to correct , and esteem them.-He that will not be corrected , or advised , in his writings , is akin to a pedant .
' ' Mons. De la Bruyere , transl .
by N. Row , Esq ; Quid dignum tanto feret hic promissor hiatu ?
HOR .
de Art .
Poet .
Reverend Sir , 1760 .
ASI highly honour your ingenuity , and applaud your candid treatment of those , who entertain sentiments different from your Read Jan. 29 , your own , while they attempt to discover useful truth , I presume to lay before you , a great master of the subject on which I write , some hints relating to what has been published , and hope to give , as has been desired ( 1 ) , a little further light , by what I have observed from my own parish for 24 years past , having noted the particular ages of 1700 persons buried in that time ( 2 ) , from the London bills , more especially on the age of 100 years , and upwards , and from the accounts of every other place I could procure .
I shall mention at present a few leading points only , and those as briefly as I can , sensible of your ability to trace their consequences , and present them as they occur to mind , on view of your table and discourse in the Philosophical Transactions , Vol. XLIX .
p. 167 .
Comparing the burials of London and Breslau , you say 8110 die at London , and 202 at Breslau , under 2 years of age .
I acknowledge it not only yours , but a current opinion , taken as a first principle , that at Breslau about 1/ 5 of those that are born die under that age ; and the place has been celebrated for its healthfulness , for the successful care of infants in particular , and for the good constitution and longevity of its inhabitants in general , a place much envied , and much contended for .
Now I grant , that the numbers 145 and 57 make 202 ; and that 1000 202 = 798 , which stand in the third ( 1 ) Richards ' Annuities on Lives , 1739 , pref .
iii .
Dr. Brakenridge , Phil. Trans. Vol. XLIX .
p. 172 , ( 2 ) Intending , if I should live , to publish them , with extracts from the registers for 200 years , and the result of the number of inhabitants twice taken from house to house .
year in Dr. Halley 's table ; but still can by no means admit the inference ( and am surprised it should ever be imagined , that of 1000 children born at Breslau , 202 only die under 2 years of age ) , for very good reasons , to be found in the book of nature , and in Dr. Halley 's dissertation too , as I shall make appear by and by .
In the mean time , let me observe how much it were to be wished , that all , who write upon this subject , would begin from the birth , or 0 year , and give a true annual register of the growing , the most confirmed , and the declining , state of life , by some method devised to make it visible at once , as I shall show hereafter , without leaving the reader to try the numbers singly upon every occasion .
For while some account from the quick conceptions ( 3 ) , as in the London bills ; others from the living births , as is your way , and I think the best ; others from 6 or 9 months , or a full year after the birth , as Dr. Halley and Mr. Kerseboom have done , ( and great is the transition from 0 year , or the birth , to those of 1 year old ) , there must arise confusion at first setting out , and apparent , if not real contradictions , in comparing one account with another .
And it should likewise be well remembered , that if a less number are taken for the deaths in the first stage of life than there ought to be , the more in course are thrown ( 4 ) back upon the ensuing decads of years ; and thus a whole table may be effected by the first year only ( 3 ) Graunt 's Observations on the Bills of Mortality , 5th edit .
1676 , p. 22 .
84 , 85 .
and from hence abortives and stillborn are included in the burials .
( 4 ) See at the end of this Letter , p. 69 .
( c ) , ( d ) , ( e ) , and ( f ) .
unfairly represented , as Dr. Halley 's certainly is , if he ever meant , as he is generally , but I think not rightly , understood .
I said I could not admit , that of 1000 children born at Breslau 202 only died under 2 years of age ; and , having prepared you for it , shall now give you the reason .
Dr. Halley states the births ( 5 ) certain at 1238 , and adds farther , that 348 of them do die in the first year ; that but 890 arrive at a full year 's age ; and that 198 more do die in the 5 years next following .
Upon which data , I state those 6 years of the local lives and deaths , and their proportionals for 1000 births , as follows .
Year begin .
Born .
Die .
Born .
Die .
1 1238 348 of 1000 281 2 890 76 719 61 3 814 49 658 40 + 4 735 35 198 618 28 160 5 730 23 590 19 + 6 707 15 571 12 7 692 , as in the table ; but 559 only , if begin with 1000 births .
+ signifies an unit taken in to the integers from a decimal of.5 or more .
And from hence I think it very plain , that out of 1238 births at Breslau , 424 die under 2 years old ; and therefore , out of 1000 , 342 die under that age : which is somewhat more than I observe in my parish , ( 5 ) Lowthorp 's Abridgment , 5th edit .
1749 , Vol. III .
p. 669 .
though mine exceeds Mr. Kerseboom 's ( 6 ) account ; and I am better pleased to find , that what I take for the truth lies between them .
And here I shall venture to assure you , upon the whole , that when brought to the due proportionals for 1000 births only , the account of Breslau falls in between ( 7 ) those of my parish and of All-Saints in Northampton ; so that in reality , there is nothing either remarkably ( 8 ) healthful , or long-lived , in the inhabitants of Breslau , as has been imagined , by mistaking our author 's meaning , who perhaps might intend his ages should imply the middle of every year , his title being age current , and from the 1238 threw off 238 only for 6 months , or more , at first setting out , and not the whole 348 , as he would have done , if intended for the whole year .
However , let his design be what it will , the number 692 ( being the remainder of 1238 after 6 years deaths ) is placed in his table ; and if we take the 692 , and what follows thenceforth , we must not forget with what number the author began , nor confound his with other tables , that really begin with 1000 births , because this has 1000 persons in the first year .
It is with great regret , that I mention any blemishes in this table , so much and deservedly esteemed , which has given the lead to many others of the like Year .
Lives .
Die .
( 6 ) Mr. Kerseboom 's table begins 1 1125 50 ; but the first year from the birth is suppressed , and 0 year should be 1400 275 .
He might have said this plainly , to prevent any stumbling at the threshold .
( 7 ) See p. 69 .
( d ) , ( h ) , ( k ) , and ( l ) .
( 8 ) Ward 's Clavis Asurae , p. 111 .
Hodgson , pref .
Stonehouse , pref .
p. 7 .
kind , but I cannot say it is exact .
For by departing from his preliminary ( 9 ) discourse in the first 6 years , and varying his table from the stated proportions , at their respective ages therein set forth , it is certain the one does not agree with the other , till the years 57 to 62 , and in the single years 72 and 80 .
And it were much to be wished , that Dr. Newman 's papers were made public , to discover the origin and justness of those proportional numbers , how they prove from year to year , and from what leading proportions they were deduced .
However , if the consideration of the first year was the only reason for desiring them , it might be needless , there being another passage in the same dissertation , which further confirms what I assert ; where , speaking of these tables of Dr. Newman , about to be ranged in particular form by himself , he makes this observation ( 10 ) .
It appears hereby , that the one half of those that are born , die in 17 years time , 1238 [ births ] being , in that time , reduced to 616 .
For we cannot , by any means , apply these words to the number 1000 in his own table , which are not reduced to half , till near the 34th year ; an assurance to all , that know any thing of these matters , that 1000 births , and 1000 persons , the one in the beginning of life , the other in the beginning of his table , mean very different things , and that any conclusions , truly drawn from the one , may be very false when transferred to the other .
( 9 ) A general state of these may be seen , p. 69 .
a yearly one would take up too much room here , though I have it before me .
( 10 ) Lowthorp 's Abridgment , Vol. III .
p. 677 .
If some mistakes , and unfair comparisons , arise from divers tables not beginning all from the same point ; if in the London accounts the abortive and stilborn must be thrown out , and also an allowance made for as many as die in the greater part of the first year ( while the 1238 Breslau births were reducing to 1000 persons only ) , before a just comparison can be made with Dr. Halley 's table , as it now stands ; then the 8110 at London ( 11 ) must be reduced , or else the 202 , or rather the 342 abovementioned , be further increased by an allowance for such abortive and stilborn .
And when these alterations are made , the accounts of the two places will not appear so amazingly different , in proportion to their respective numbers , upon the whole .
Beside the inconvenience of the various accounts not beginning together , I shall add another objection no less material ; that the tables are formed in too small numbers , and , by that means , cut off 20 or 30 years of the term of life , and undervalue it in annuities , as nothing worth .
He that begins with 1000 only , either stops short of 90 , or runs quite out between the 95th and 100th year , and can go no further ; because out of 1000 births , it is not expected , so much as one should arrive at the age of 100 years .
But what must become of those many ( 11 ) Adding 2000 to the burials divided by 10 , is , in effect , adding 20,000 to the whole .
The increased column is 5 short , which would arise from additional parts lost , and make the full sum .
The last number in the Breslau column should not be 33 , but 27 .
Dr. Brakenridge .
in the London ( 12 ) accounts ( for instance , in the 30 years , 1728 to 1757 inclusive ) 2979 living at 90 , 242 living at 100 , 10 living at 110 , and one living to 138 ?
Are these to be wholly omitted by those , who pretend to give a true state of human life , the first number near thrice as many as the usual tables begin with ?
Or , could they be overlooked , if the computation began with 100,000 , or a million ?
Further , should the value of annuities sink so precipitately , and close so soon , will they be granted to persons aged 95 for nothing , as the table of Mr. de Parcieux has it , in the supplement to Chambers 's Dictionary ?
One would imagine thence , that those aged 100 , or more , should have a premium to accept of them .
And yet , what would be the consequence , if the state granted an annuity to 100,000 persons , and the survivors of them , to subsist entire to the death of the longest liver , and have it to pay 20 or 30 years beyond the utmost expectation represented in such tables ?
It may here be observed , in respect to London more particularly , that the induction from this residuary part of life is well supported , since at 90 , or later , few think of removing from town .
Or , if it be insisted upon , that some do remove after that age , it will be allowed , that the burials of the subsequent years would have been higher , if they had all staid and died there .
It may seem quite impertinent to mention this to ( 12 ) Mr. Stonehouse forms his account on 529,623 , of which 181 lived to 100 years , and upwards ; yet , beginning with 1000 , closes his table at 95 .
It is needless to name many others in a general fault , to which the easy management of small numbers is a temptation .
you , who have assured us , in general terms , that the burials ( 13 ) after 50 are less than they ought to be , allowing the accounts sufficiently correct , and the numbers in later life not exaggerated , but rather the contrary .
It will be found then of particular service , that for those of 100 years old , and upwards , we have the age of every single death ; and forming a table of them yearly decreasing , and applying Dr. Halley 's third rule of halving the tabular lives in any year , to discover the term expected , it will come out , that a life being , like one of the 242 aged 100 , has an equal chance to live 2 years 3 quarters , or more ( 14 ) ; and , by his process for finding the value from yearly chances , and at 5 per cent. it amounts to more than two years and a half purchase ( 15 ) .
Now , by your table , a life of 85 has not a better expectation ; and following too close upon your heels , Mr. Dodson values an annuity of 1 l. for a life ( 16 ) of 88 but at 6 s. and 5 d. ready money .
I shall not controvert this point ; but desire to know , who will grant such annuities , or greater , for all that could be found of that age , or as many of them as should be selected for nominees : I say nothing of the first number in his table .
Doubtless you mean by 1 the first year of new-born children ; and yet if he means the same , ( 13 ) Phil. Trans. Vol. XLIX .
p. 175 .
( 14 ) As all the tables do or would reduce life to 0 before 100 , how will they emerge again , to join conformably with these in term expected , or value ?
Yet these are realities , set in public view .
( 15 ) Process in MS .
fol. 4 .
( 16 ) Phil. Trans. Vol. XLIX .
p. 891 .
and values an annuity for a new-born child , whose equal chance of life is not 4 years , at 12.51 years purchase , it appears to me greatly over-rated , and I should suspect some fallacy in the method of computation .
Another thing I shall propose to your consideration , is the forming a continual register , if I may use that term , of the proportions of lives and deaths , by adding after each year , how many would have died out of 1000 , or one out of how many , or both of them , in subsequent columns , with the differences , increasing , or decreasing , from year to year .
If one only be used , I rather prefer the former , as it strikes the eye , is a more natural representation of increasing mortality , and shows at once , if 1000 were to begin every single year , how many of them would die in that year , in proportion to the lives and deaths of such year in the table .
The latter may be more agreeable to others , and is of singular use towards the end of life .
I shall give a specimen of both on your table for 3 decads , by which you will better perceive what I mean , and the uses ( 17 ) that may be made of them .
( 17 ) Vide p. 61 .
Year .
Per One out Year .
Per One Year .
Per One out Lives .
Die .
mill .
of Lives .
Die .
mill .
Diff. out of Diff. Lives .
Die .
mill .
of -.12 1 .
11 406 4 9.85 101.5 21 368 4 10.87 92 .
31 325 5 15.38 65 .
-.12 1 .
12 402 4 9.95 100.5 22 364 4 10.99 91 .
32 320 5 15.63 64 .
-.12 1 .
13 398 4 10.05 99.5 23 360 4 11.11 90 .
33 315 5 15.87 63 .
-.13 1 .
14 394 4 10.15 98.5 24 356 4 11.24 89 .
34 310 5 16.13 62 .
-.12 1 .
15 390 4 10.26 97.5 25 352 4 11.36 88 .
35 305 6 19.67 50.833 -.13 1 .
16 386 3 7.77 128.66 26 348 4 11.49 87 .
36 299 6 20.07 49.833 -.14 1 .
17 383 4 10.44 95.75 27 344 4 11.63 86 .
37 293 5 17.06 58.6 3.08 18 .
18 379 4 10.55 94.75 28 340 5 14.71 68 .
38 288 6 20.83 48 .
-.22 1 .
19 375 3 8.00 125 .
29 335 5 14.93 67 .
39 282 5 17.73 56.4 -.22 1 .
20 372 4 10.75 93 .
30 330 5 15.15 66 .
40 277 6 21.66 46.166 3885 38 97.77 1034.66 3501 43 123.48 824 .
3014 54 179.63 563.832 It is usual with me , to note the yearly differences of mortality , which I could not do here , on account of the iregularity , but only in decade 21 , &amp;c. By the way , I must give a reason why I sum up the column of lives , which I consider as so many annual exposures , and this as the total of the lives , each exposed to the chance of mortality for one year ; ( i. e. ) 406 in the first , 402 in the second , &amp;c. and 3885 in the ten years ; and , upon the whole , 38 deaths .
Exp. Die .
Exp. Die .
And thence 3885 : 38 : : 10,000 : 97.812 ; which last term is the proper state or degree of mortality for that decade , and 9.78 a mean thereof , at an average .
It is generally acknowledged , that some one between the 10th and 20th is the healthiest year , i. e. the year in which fewest would die out of 1000 , and the annual degree ( 18 ) of mortality should increase ( swifter or slower as it happens ) from thence to the end of life .
But how is such year to be found among the irregularities of the first of these three decads ?
Or how shall we look upon 10 .
11 .
14 .
15 as a due progression in the second ?
And if the numbers 16 .
19 .
20 do go on increasing in the third , why does the degree of mortality go back to 17 , then forward to 20 , then back to 17 again , and forward to 21 per mill ?
in such a manner , that one out of 49.8 should die at the age of 36 , and but one out of 58.6 at the age of 37 ; and again , one out of 48 should die at the age of 38 , and but one out of 56.4 at the age of 39 ?
Is not this representing the 37th ( 18 ) A mistaken inference from this , see p. 65 .
note ( 30 ) .
year of life as a healthier than the 36th ; and the 39th healthier than the 38th year of life ?
I am sensible you might say , this is owing to the promiscuous changes of 5 and 6 in the deaths , for which you may have reasons , though I cannot fathom them .
With me it is not a matter peculiar to your table , but a certain consequence of beginning with 1000 only : for having no changes , but what amount to an unit more , or an unit less , that is too great a leap at once , in such small numbers ; when there might have been , in 500 and 600 deaths , room to express duly the intermediate gradations of increase , or decrease , for every single year .
The last thing I shall mention , is the term or expectation of life , shortning too swiftly , and then recoiling .
I shall instance in your table , and Dr. Halley 's too ( and the same will be found in the rest ) , and apply to them his rule above-mentioned , and the term , or probable expectation of life , will come out thus , for the ages following .
Year .
Lives .
Die .
Per One Term Year .
Lives .
Die .
Per One Term mill .
Diff. out of Diff. of life .
Diff. mill .
Diff. out of Diff. of life .
Diff. 77 43 6 139.5 7.16 4.12 77 68 10 147.0 6.8 4 .
3.26 1.00 -.25 8.1 -.36 -.17 78 37 6 162.1 6.16 3.87 78 58 9 155.1 6.44 3.83 -.04 8.1 -.32 -.13 79 31 5 161.2 6.2 3.83 79 49 8 163.2 6.12 3.70 7.5 -.27 80 26 4 153.8 6.5 4 .
80 41 7 170.7 5.85 3.83 1.97 -.66 5.7 -.19 81 22 4 181.8 5.5 4 .
81 34 6 176.4 5.66 4.50 2.1 -.06 82 18 3 166.6 6 .
4 .
82 28 5 178.5 5.6 5 .
-.08 83 15 2 133.3 7.5 3.75 83 23 3 130.4 7.66 5.25 -.50 84 13 2 153.8 6.5 3.25 84 20 2 100.0 10 .
5 .
-.00 -.00 -.50 85 11 2 181.8 5.5 2.75 85 18 2 111.1 9 .
4.50 According to Dr. Brakenridge 's table .
According to Dr. Halley 's table .
The column per mill , should increase ; and that one out of , should decrease , with some sort of regularity : I say no more of these , referring to what I have written above .
My design here is to show , that the term of life ( 19 ) decreasing too swiftly , sinks below the truth , and then stands still , or increases , to become agreeable to the rule of nature , found in the course of subsequent years .
These irregularities and disproportions might be avoided , by beginning with 43.00 and 68.00 ; or , it might be as satisfactory to many , to decrease the 4.12 years expected to 2.75 , by proper intervals , or differences , greater above than below : an expedient not to be despised , when the capital points , from and to which , are previously settled on good authority .
The term found by Dr. Halley 's table is still wider from the due course .
I am apt to imagine , the consciousness of these difficulties induced you to stop short , and perhaps you will wonder , that any one else would take the pains to surmount them .
But to return .
You may possibly think it unreasonable , that any body should insist upon such a variety of scruples , in a case where the best means of information are too lax , and general ; yet , I hope , will excuse them , when assured , that my sole reason for insisting upon them , is not a proneness to find faults , but an earnest ( you may call it an over-earnest ) desire to make what we have better understood , and attain further means of such knowlege , and state ( 19 ) I use this as the shortest form , meaning the equal chance , or probable expectation of life ; as many surviving , as dying before such period , among lives of the given age .
them in the least exceptionable manner : for till we get a more authentic account of life , or use what we have , without too much obsequiousness to great names , we shall be as if blind and fettered .
During such prejudices and restraint , it is too early to compute values with minute precision , as some do , and wrangle about trifles , while they suffer first principles of greater moment to pass quite unregarded .
Whether my conjecture be right or not , Dr. Halley had reasons why he left off at 84 , as you may have for ending at 87 ; and the term of final direction is settled by another great master ( 20 ) at 86 .
But when many persons outlive such tables , and are most desirous to purchase annuities , upon easy terms , for their lives , and have no rule at all left , it must be very acceptable , by whomsoever faithfully performed , to have a table beginning with the living births , formed upon 100,000 lives at least , and carried on to the extremity , I should almost say the utmost possibility of life , with the swift or slow increase of annual mortality , noted in a subsequent column , and in consequence the term or expectation properly decreasing , from the best life about 5 or 6 , till the whole be exhausted : and it would be a satisfaction to me , if , by suggesting any hints , it may put abler hands at work , to bring it at last to such a perfect state , as I conceive , at present , in imagination only .
In prospect of this , give me leave to observe , that the numbers in those columns Per mill and One out ( 20 ) Mr. De Moivre , pref .
p. v. ( edit .
1725 .
) tract , p. 10 .
47 .
76 .
79 .
This strikes off 14 years of Dr. Halley 's table .
Younger lives can hardly look so far forward , but old persons see them at hand , and the value of all expectations is in proportion to their near approach .
of , may be of use for such a purpose , as directors ; and , by inverting the proportionals , be applied to find the deaths of lives given , from year to year ( 21 ) : for as these may be previously digested , with greater regularity , and the number of deaths found , by multiplying the given lives by the former , or dividing them by the latter , a way is opened for conducting the work in larger numbers , and with great exactness : and I conceive , that ten of the best accounts of different places , each formed by proportion for 10,000 births , and all thrown together in the usual sections , if properly aided in the latter part of life by the London bills , might be sufficient , at least , would be much better than any thing we have ; for tables of 1000 , 1238 , or 1400 , are quite distanced here , having no lives at all of 100 years .
And yet , who can believe that Dr. Halley 's 34,000 , if they were so many , much less Mr. Kerseboom 's 980,000 , had none of 100 years and upwards ; when the London bills afford about 8 in a year ; and , upon the whole of 750,222 burials in the said 30 years , have 242 of that age ?
Or , who can give a reason , if they had such , why they constructed their tables on so small a basis , as wholly to exclude them ?
It may be said , the 242 persons , dying above 100 years of age , did not arise from 750,222 births in town , but from a great many more brought in from ( 21 ) These might be tried , or regulated , by proportions formed from other accounts , taking the deaths between 10 and 20 ( or the most certain period ) in each , for the two leading proportionals , and trying above and below for lives and deaths .
Lesser accounts of single parishes will not do alone ; for as aged lives are rare , one is forgotten before another happens , and in small numbers they are not to be expected .
the country after 20 , and probably , upon the whole , from double that number of births .
I shall only reply , let the number of births or persons , natives or aliens , be what it will , from which the 242 arise , yet , in some definite number , such and so many instances of longevity are found , of which the present tables take no notice ; and though the fact is manifest enough , yet the absurdity , in respect to practice and formation of tables , still continues .
If we look back , we shall find the first sketch , that of Capt. John ( 22 ) Grant ( alias Sir William ( 23 ) Petty ) was formed upon 100 only , and such a table carried the account to the 80th year , or upwards .
Next were introduced those of 1000 , and extended the computation of life to between 84 and 100 ; tables formed upon 10,000 would advance to above 105 ; and upon 100,000 , duly proportioned from the materials we have , might continue the account to 115 years , and upwards .
If in the first sketch , the supposed term of life was closed too soon , and it was an improvement to carry on an account of the gradual decay beyond the 90th year , why are we to rest here , having additional observations made for more than 60 years , which furnish materials for a further progress ?
If there is room , and good foundation to advance but 20 years beyond the compass of the present tables , should not this be done ?
And will it not make a considerable , yet necessary , alteration in all computed values , upon annuities to be granted to persons in the latter part of life ?
( 22 ) Graunt 's Observations , 5th edit .
p. 84 .
( 23 ) Phil. Trans. No 196 .
I have not seen Mr. Smart 's tract on the London bills , ( when and how was it published ?
) nor the collections of Mr. Dupre , published by Mr. Buffon , save only through the medium of Mr. Kerseboom 's proportions ( 24 ) , beginning with 1000 , not births , but children of 6 months old , or upwards ( 25 ) ; which makes a considerable difference in respect to age given , and yearly deaths , through the whole table ; and I almost envy those , who have the desirable use of choice originals .
The accounts of Dr. Newman are , I suppose , preserved by your society ; and there is a state of the exchequer annuitants ( 26 ) often mentioned , but not published , by Mr. Lee .
Nevertheless , these last being of divers ages ( if the particular age of each person at entry and death be not known , though the gross numbers yearly dying may ) , as it was too great a presumption to assert , that they began all at the best stage ( 27 ) of life , and were so nicely chosen ( 28 ) , that the duration of 35 years was a thing extraordinary ; so it would be a blameable credulity to admit these points for truths , when we continually see how many are resolved to chuse their own lives , or those of their children or favourites , even when they are receded 10 or 20 years from that part of life , which had the largest expectation .
Whether it was this matter better con(24 ) Phil. Trans. 1753 .
p. 239 .
( 25 ) This was done , to compare it rightly with Dr. Halley 's , which Mr. K. therefore knew was not from 1000 births .
( 26 ) Lee 's Essay , 252 , 253 .
` ` This , ' ' he says , ` ` is the best guide of all .
' ' Lee 's Val .
Annuit .
p. 47 .
51 .
( 27 ) Essay 252 .
( 28 ) Essay 253 .
sidered , or whatever else changed that author 's sentiments , yet changed they were ; for , in 1737 , he accounted a life of 10 years best , and equal to a term of 28 years , and no more ; [ Lee 's Essay , p. 231 .
253 .
] and yet , since , in ( his Valuation of Annuities , 2d edit .
p. 96 .
) 1754 , he has computed the same kind of life , as equal to a term of 35 years , notwithstanding all the allowances pleaded for in his Essay ( 29 ) ; in full consideration of which , he was afraid of overdoing the matter .
And yet , if he would have given us the true result of the London bills , according to his own state , and reckoning with exactness , which he calls to the extremity , a life of 10 years would be found equal to a term of 34.94 years ; but one of 4 years old equal to a term of 38.20 years ; above 3 years better than his best life .
I should not have mentioned this , but to observe , that it is natural enough , when the expected term of life is taken to the uttermost , to make some allowance .
And yet , if allowance is to be made , it should not be by an arbitrary and false representation of life throughout a whole table ( 30 ) , but left to the discretion of parties concerned , and to be made in proportion to ( 29 ) Page 231 .
( 30 ) Mr. Lee rightly concluding , that the degree of mortality ought to increase from the year of greatest expectation to the end of life , erroneously inferred from thence , that the deducend , or part of the term of life wasting in each decade , should be least at first , and greater afterwards ; and so apportioned it , as long as he could go on , contrary to all other tables , and even to the course that would arise from his own table of the London bills .
Lee 's VO L. LII .
K the Essay , p. 459 .
Table II .
the term found by such table ; and that , in general , about 1/ 10 of the term deducted , in any part of life , would not be unreasonable , when the account is carried on to the full extent of human life , especially if such supposed duration is taken as a rule for price ; for reasons well known to you , by comparing the respective ( 31 ) value of present and future years at any rate of interest ( 32 ) , and observing , that a gain of time in the latter , would not be equal in value to the loss of like time in the former .
After many degrading comparisons above-mentioned , to the dishonour of our capital , it may counterbalance them , to hear what this advocate has urged for its healthfulness and longevity , on his own experience .
( Lee 's Essay , p. 252 , 253 .
) I do not intend to disparage it , by mentioning a place I have known above 40 years , never famed for salubrity , and yet has produced as many ( may I say more ?
) persons of 90 , and upwards , than London , as would appear on a fair proportion , formed on the burials of each , by a method too long to be laid before you at present ; and I am ready to oppose this , as I did to the former complaints , so to these boastings , being both alike ill-grounded and unreasonable .
As my objections extend to all the accounts I have met with ( perhaps I should except Mr. Dupre 's , if , I could see it entire ) , you will naturally expect what ( 31 ) This , I find , has been noted by H. B. in observation on Lee 's Essay , 1739 .
p. 18 .
21 .
33 .
( 32 ) This makes hazard of time and hazard of value unequal in course .
I have to offer more perfect in their stead .
The ground-work I present would lie in a nut-shell ( 33 ) ; the edifice would supply matter for such another letter : and , I think , you will be glad to have a little respite , to consider of this , and judge how far some facts herein set forth ought to be regarded , without which , I presume to affirm , all computations of value will be found inconsistent , and very faulty in some other parts of life .
Upon such an occasion as this , I hope you will excuse the intrusion and tediousness of a new visitor , who aims at further improvements on this subject , both in matter and form .
I willingly submit the result of not a little time and pains to your superior judgment , and beg you will either frankly pronounce it time misspent , and labour in vain , or else , by your kind directions , enable me ( if the ardor of present inclination should continue ) to improve and finish a scheme , perfect enough to bear public view .
I can only say , I should endeavour to keep clear of the objections made to other tables , and to support whatever I advance by real facts , or very strong probabilities , and to make the whole consistent and uniform .
I do not enter upon any computation of values , nor prefer either of the two methods ( 34 ) for finding them ; let the facts , upon which they depend , be better ascertained .
In order to this , the sole object ( 33 ) See p. 69 .
letters ( g ) , ( h ) .
( 34 ) One proposed by Dr. Halley , and approved by Mr. De Moivre , the other by Mr. Lee .
Richards 's Annuities , p. 1 .
Lee 's Valuation , p. 2 .
of my present view is , to state , with all exactness , the term and period of human life , being ambitious so to execute this underpart of the work , as may deserve yours , or general approbation .
Having delivered this message , you may form in idea the aukward rusticity of a stranger , introduced the first time to your presence , who hardly knows how , or when it is fit , to make his bow , and withdraw .
I am , Reverend Sir , Your great admirer , and most respectful humble servant , T. W. N. B. In what follows , I am obliged to number 1 year beginning , for the sake of others , which I usually mark 0 , and the rest 10 , 20 , 30 , &amp;c. POSTSCRIP T. Year beginning .
Deaths .
Deaths .
Deaths .
Deaths .
Deaths .
Deaths .
1 5791/ 2 495 581 469 347 122 less .
806 474 667 459 4661/ 2 Thrown back upon these decads 11 46 39 62 50 61 11 93 55 + 61 42 + 481/ 2 21 68 58 68 55 69 14 114 67 130 90 + 781/ 2 31 81 69 861/ 2 70 87 17 138 81 96 ( 1 ) 66 731/ 2 41 84 72 100 81 101 20 130 ( 1 ) 77 + 105 72 741/ 2 51 1001/ 2 86 1031/ 2 84 103 19 148 87 114 78 821/ 2 61 103 88 1001/ 2 81 101 20 135 79 119 82 + 801/ 2 71 85 73 99 80 97 17 81 48 + 97 67 + 571/ 2 81 19 16 30 24 27 3 44 26 + 58 40 + 33 91 41/ 2 4 71/ 2 67 ( 3 ) 1 11 653 41/ 2 101 0000000011+ ( 2 ) 01/ 2 11701/ 2 1000 1238 1000 1000 122 more .
1700 1000 1453 1000 1000 ( a ) ( b ) ( c ) ( d ) ( e ) ( f ) ( g ) ( h ) ( i ) ( k ) ( l ) ( 1 ) The case of a lesser decade between two greater , is not unusual .
I find the like in Dupre , Kerseboom , Edenborough , and Norwich , not all in the same decade : divers accounts rectify one another ; a single one may be made discreetly smooth .
( 2 ) At ElyTrinity , one now ( 1760 ) living aged 103 , and one died within memory at 106 , and upwards , may be set at about one in 3000 dying at 100 years old , or upwards ; but who will live to take account of them ?
( 3 ) If ( e ) were taken unreformed , Breslau would , after 10 , be more unhealthy than ( h ) or ( k ) .
( a ) The result of Dr. Halley 's proportionals , ( Lowthorp , Vol. III .
p. 670 .
) filled up from year to year , to the 100th year .
As these come so near 1174 ( stated p. 669 .
as the annual deaths ) , it should seem as if the first attempt was made that way , to find how many of such deaths would fall in each decade .
( b ) Exhibits the same for 1000 births .
( c ) By taking in the 238 that were suppressed , stating the first 6 years , according to the preliminary discourse , and continuing the account according to the table ; only setting back half a year , or postponing 4 deaths to the 2d decade , [ and so half the last year 's deaths from every decad to the following ] and this gives the 1238 entire .
( d ) Exhibits the last preceding , computed for 1000 births , and as the table ought to be from those data , but yet ill proportioned .
( e ) Shows the decrements , or deaths , according to Dr. Halley 's table , beginning with 1000 persons , and exhibiting a supposed body of coexisting people , in all 34,000 .
The first decade , so much below the truth , from 1000 births only , that it swells the 9 decads following with 122 deaths more than should be , as is represented in ( f ) .
( g ) The deaths of 1700 persons in the parish of Ely Trinity .
( h ) The same proportional for 1000 births .
( i ) and ( k ) The like for All-Saints in Northampton .
( l ) The halves of the sums of ( h ) and ( k ) for 1000 births .
For , by comparing ( d ) and ( e ) , you will see , that the first decade being less than it should be , ( i. e. ) so much short of 469 , throws back 122 deaths upon the other decads of life .
And if this were 469 , as it ought , it would fall between ( h ) and ( k ) , Ely-Trinity and AllSaints in Northampton .
As this happens in the first decade , I have joined both accounts , and halved them ; and the result of it is stated at ( l ) ; which will be found as near ( d ) as can well be expected , in accounts of different places ; and with this I shall conclude this Postscript .

