( Kapital-i, in 'I, myself', now spels Il as in isle or aisle.
Leter y spels sym for seem or seam and partys for parties.
Leter w spels swn for soon. )
Links to sektions:
The next sektions folow the steps to proportional representation, yusing the Senatorial Ruls of transferabl voting:
I rylIs'd that sientifik method kud solv the problem of voting method. But such a formal aproch may only apyl to specialists and not the jeneral publik. As a student, hw hadnt yet voted, I thot ther was nothing to no about voting, apart from plasing an X. Beginers dont no they ar beginers, and ar not hapy about finding it out. Perhaps, the best way to begin with them is at the begining of elektoral reform.
John Dewey's greit edukational reform slogan was 'Lerning By Doing',
bekaus that was much mor efektiv than kraming children with
akademik abstraktions.
In 1821, Thomas Wright Hill sujested the prinsipl of proportional representation
as it was informaly praktis'd in his boys skwl. To quot Enid Lakeman, in How
Democracies Vote:
...his pupils were asked to elect a committee by standing beside the boy they liked best. This first produced a number of unequal groups, but soon the boys in the largest groups came to the conclusion that not all of them were actually necessary for the election of their favourite and some moved on to help another candidate, while on the other hand the few supporters of an unpopular boy gave him up as hopeless and transferred themselves to the candidate they considered the next best. The final result was that a number of candidates equal to the number required for the committee were each surrounded by the same number of supporters, with only two or three boys left over who were dissatisfied with all those elected. This is an admirable example of the use of STV.
It is yusful to no that som kandidat is a klyr favorit, bekaus he may be mayd the chair-man. Other posts on the komity may go to thos next most favor'd with mor vots than they nyded to be elekted.
But Hill's esential idea is of transfering surpluses from thos hw hav mor to thos hw hav les than they nyd. And it may hav byn the moral to the Gospel insident of the louvs and the fishes. Jesus had his disipls go round an audiens to gather the left-overs for thos hw had stil not byn fed.
Thomas Hill's orijinal proportional representation, by transferabl
voting, is stil the best way to grasp the prinsipl. Its virtu is that it is jenuinly
demokratik. No-uon tels the boys hw to suport. No masters or prefekts draw up 'party
lists' to draft the boys into skwl 'houses' or tyms or wat-ever.
The byuty of Hill's elektion is that the voters no intuitivly how it works. Striktly
spyking, nobody has to do any arithmetik.
No balot papers hav to be fil'd: A voter dosnt hav to nout down that first ( or number
1 on a balot paper ) he or she stud by the skwl favorit, hw didnt nyd the suport; and that in the sekond plas ( number 2 ) she mov'd to a kandidat, hw had help'd her with her problems, but hw was also exyding the sher of vots nyded for a komity syt; the voter in question perhaps finds that by moving on a third tIm ( wich wud be number 3 on a balot paper ) she helps a personal frend to just rych the portion of vots to ensur a syt,
and so stays in the qu of her third chois.
So, this informal elektion dos not requir voters to be numerat. They dont hav to kount up to thry or fIv. But by moving on, or transfering themselvs, as surplus voters, from uon kandidat's qu to an other, they ar just as shurly expresing an order of preferens, by voting with ther fyt.
Also, a returning ofiser, vital to a formal elektion, is not nyded. No-uon has to do sums to show
a kandidat has over haf the voters nyded to win a singl syt. If 16 children lIn up behind
uon of ther tw favorits, and uon qu is 9 children long and the other 7, yu no that
the kandidat with tw extra kids has won. Yu dont hav to kount the tw rows.
If both rows ar the saim lenth ( 8 kids ych ) then lots ar drawn or yu tos a koin
to desId wich kandidat wins.
If ther ar tw syts, the winers nyd uon-third of the vots ych. Supos ther ar 24 voters. Provided tw of the kandidats both kolekt rows of 8 children behind them, they kan not be byten by any other kandidat. It is just posibl that a third kandidat may muster the remaining 8 children.
This wud be an unusual alInment. Probably, it wud be evident, wich of the thry kandidats was a hom of last resort, by having his qu of 8 komplyted last.
If not, uon of the thry tI'd kandidats, hw drw the short straw, wud step down.
( If the result was stil in disput, then the arithmetik method of transferabl voting, deskrib'd below, mIt hav to be yus'd. For praktikal purposes, that has sufIs'd. Transferabl voting dosnt hav to be perfekt to be very much on the rIt lIns. Thos hw demand perfektion in this method usualy want som other voting system very much wors, insted. )
Elekting a komity of thry, we folow on from the abov exampls of singl-syt and dubl-syt
elektions. The thry winning kandidats, hw hav uon-quarter of the vots ych, kan not be
byten. Again, the wining proportion ( or the 'quota' ) is 8 vots ych, if ther ar 32 voters, this tIm.
A 4-way tI is very unlIkly. Any tI is mor unlIkly in an informal elektion than a formal uon. Bekaus, the voters, themselvs, sying a tI, ar lIkly to brek ranks just
enuf to provid the tI-brek.
The formal elektion may gain in presision, but lak the informal elektion's ability to modify the result, under sirkumstanses that kud persuad som voters to revis ther desisions.
Nor dos it mater if thry rows of 9 children form to ensur the thry most popular kandidats elekted. Al that myns is the wining kandidats hav surpas'd the lowest hurdl of popularity, of just 8 vots ych, nyded for elektion - and wich ensurs the minimal proportional representation of thry-quarters ( or 3 x 8 = 24 out of 32 ) of the voters.
Remember, I hav to deskrib this informal elektion to yu, in terms of arithmetik, but the children elekt thry kandidats in terms of the thry longest qus. This is no dout a sofistikated proses, involving sutl jujments of the karakters of ther skwl felows, but it is not formal arithmetik. Such powers of mental arithmetik, that individuals may bring to ther aid, avoid the tedium of working with abstrakt numbers. The children ar the numbers. They perform a kolektiv kalkulation by the operations of moving about betwyn the kandidats.
The Amerikan sivik organisation 'Democracy 2000' kondukted desision-making work-shops both in skwls and town hals. They found that sitizens dont tak to the formal arithmetik of transferabl voting that the returning ofiser has to do. For beginers, I rekomend Thomas Hill's method to show a smal group, in praktis, the prinsipl of demokratik elektions.
But on a larj skail, thys elektions hav to be organis'd on formal lIns, with balot papers and a results shyt, rekording the stajes of the kount - even if the houl proses is automated. With this in mind, I'v also rekomended the ( traditional ) Irish version of singl transferabl vot.
It alow'd the formal elektoral prosedur of transferabl voting to be syn in terms of individual voters ( akording to ther orders of chois ) alokated to individual representativs making komunity desisions.
Demokrasy has byn so abstrakted and remov'd from every-day konserns that any number of bad systems ar yus'd to the diskredit of its naim. And fw pepl sym any the wIser.
Modelers of grup desision-making, lIk Democracy 2000, requir the voter to be represented by a lejislator of his or
her chois. They found the Senatorial ruls of surplus vot transfer ( a mor
advans'd form of STV, explain'd below in suksyding sektions )
hard to explain to Amerikan audienses.
( Many pepl dont hav a hed for figurs or hav a fobia of them. )
The senatorial ruls ar konsistent with uon person, uon vot ( and uon vot uon valu ). But demonstration elektions may want to styr klyr of the ambiguity of voters as mor or les represented by ther ranj of preferenses.
The old Irish system of STV, adapted for audiens inter-aktion, is rufly as folows.
This akount also servs as an atempt to show that STV kan be fairly simply explain'd.
We supos the lokal publik invited to an informal myting, so balot papers or macyns ar optional. But for the elektion of any given komity, the audiens nyd to no how many syts ther ar, and how many pepl it taks to elekt ych representativ.
For this purpos, it is explain'd how the Droop quota folows from the singl majority system. Haf the voters kan elekt uon kandidat. ( If tw kandidats ych got 50 out of 100 vots, that wud be a tI brek and they'd hav to draw straws. )
Folowing from that, tw syts kan be won by kandidats hw hav ych won uon-third the vots. ( For exampl, wen tw kandidats hav ych got 34 vots out of 100, ther ar only 32 vots left and no remaining kandidat kan win a syt with les than the requir'd proportion of uon-third the vots. LIkwIs thry komity syts ar won wen thry kandidats hav ych rych'd the quota of uon-quarter the vots. And so on.
AlrIt, we asum yor audiens hav got to no the kandidats and they ych gather round ther most prefer'd kandidat. Usualy, som kandidats wil be mor popular than others. Say, ther ar haf a dozen kandidats kompyting for thry syts. With 100 voters, elektion depends on wining uon-quarter of the vots or 25 vots. But the most popular kandidat may hav 40 pepl around her, hw lIk her best.
She dosnt nyd 15 of thos vots to elekt her as the 'best' kandidat. Thos 15 vots ar wasted unles they help elekt ther next best chois. Irish-stIl elektions wud pik 15 of the 40 at random, so they ar a representativ sampl of al the voters hw gav ther first preferenses to the lyding lady.
( The orijinal Irish method is a statistikal aproximation of the Senatorial ruls - explain'd below. Ther's no nyd for yor lyding lady's 40 first preferens voters to no that the transfer valu of ther sekond preferenses ar 15/40 or 3/8 of a vot ych. )
To kut a long story short, say thos 15 vots go to a kandidat with 10 voters around him. He also is elekted with 25 vots. We'l asum an other kandidat got just 25 vots and al thry syts ar taken. How-ever, I want to point out that this method of surplus transfer by representativ sampl givs the uon-to-uon relations betwyn voters and representativs, that sivik grups, lIk Democracy 2000, requir for ther demokratik work-shops.
Nout that the method of transfering vots by representativ sampl also preservs the basik prinsipl of uon person uon vote.
Tho not nesesary for ther voting audiens, the organisers, as part of ther studys, wud rekord the stajes of the kount - just lIk returning ofisers in ful skail Irish jeneral elektions - wich ar kynly folow'd by the publik, hw voted in tw referendums to kyp STV, or 'chois voting', as it's often kal'd in Amerika.
The next sektions folow the steps to proportional representation, using the Senatorial Ruls of transferabl voting.
The kandidats' naims ar printed in random order on the voting papers. Ych kandidat may label wat they stand for. If the elektion is politikal, this may be a party label or independent. The elektorat has a preferens vot. Ych voter may stait ther order of chois for the kandidats, by plasing the number 1 by the kandidat's naim of ther first chois, number 2 by ther sekond chois, number 3 by ther third chois, and so on, in order of preferens, in so far as a given voter plyses.
With the STV method, later preferenses do not kount against former preferenses. A 3rd chois wil not wyken the chanses of a 2nd chois, nor a 2nd chois wyken the chanses of a 1st chois.
With STV, the voter dos not hav to vot only for kandidats of the saim party. The voter may prefer kandidats of uon party - or kandidats of tw or thry partys, perhaps to expres a wish they form a ko-alition. The voter may prefer among kandidats of uon party or of several partys, and among independents.
In fakt, the voter may stait any order of preferens betwyn al the kandidats on the voting paper. But it is not kompulsory to number-order al the kandidats.
It shud be stres'd that number'd ordering uon's chois of kandidats is al the voters nyd to no about the transferabl voting method of elektion. Every kind of elektion is in tw feizes. First, the aktual voting and sekond, the kount of the vots. The kount is kondukted by train'd personel.
The Electoral Reform Society Of Great Britain and Ireland has a rul
bwk ( as by Frank Britton and Robert Newland ) to kover every eventuality in
the prosedur of an STV kount and jeneraly organis a parliamentary
elektion down to the last detail. For a long tIm, they hav nown about
an extrymly thoro program for komputing transferabl vots.
They, or ther veteran sister sosiety, the Proportional Repesentation
Sosiety of Australia may help with specialis'd enquirys.
But the folowing information shud be enuf to elekt yor own interest
grup's komity by STV, as milions of pepl alredy do.
Hav a kalkulator handy.
The kount of the vot myns that, first, the number of voting papers is
aded up, and ych paper is chek'd to sy it has byn korektly fil'd
in, with the preferens numbers plas'd within the spases provided by the
kandidats' naims.
If, for instans, the voter has mayd a mistak by mising out the
number 3 but has plas'd numbers 1,2,4,5, then his voting intentions wil
only be kounted valid for the first tw preferenses, 1 and 2. Any invalid
voting papers are subtrakted from the total vot, to giv the total valid
vot.
With STV, kandidats ych nyd to win a proportion or quota of the total
vots befor they ar elekted. This proportion is kal'd, after its
inventor, the Droop quota.
It is a rationalisation of the simpl majority kount, wer-by uon
kandidat nyds just over haf the vots to be elekted in a singl member
konstituensy. The Droop quota jeneralises this to say that in a tw-member
konstituensy, tw kandidats nyd just over uon-third the vots
ych to be elekted. This givs a proportional representation of tw tIms
uon-third or tw-thirds of the total vot.
Thry members ar return'd with uon-quarter of the vots ych,
proportionaly representing thry-quarters of the voters. And so on.
Therfor, in a 7-member konstituensy, the PR is 7/8 of the voters.
As a simpl exampl, tak a klub of 32 voters, hw wish to elekt a komity of 3. (A gud tip is always to hav an od number of komity members to avoid posibl stail-maits.) Advisedly, the elektiv quota is simply uon-forth of 32 vots, wich equals 8 vots.
( Round-up smal fraktional quotas, e.g. 35/(5+1) = 5.84, rounded-up to tw desimal plases. )
Supos ther ar 5 kandidats, kal'd A,B,C,D and E. Having kounted 32
valid vots, the returning ofiser then kounts how many first preferenses
ych of the 5 kandidats hav. It is standard praktis to openly tabulat
the stajes of the STV kount. So, wat yu hav is a bord rais'd to publik
viw. On top, the quota, 32/(3+1) = 8, shud be noted.
The far left sId kolum randomly lists the kandidats' naims. The
stajes of the kount ar rekorded in suksesiv kolums.
Firstly, the number of first preferenses for ych kandidat ar riten in the kolum besId ther naims. Say, kandidat D has 2 first preferenses, A has 5, E has 6, C has 12 and B has 7.
( By way of kros-cheking, al thys first preferenses are toted-up. In fakt, ych suksesiv staj of the kount must kros-chek with the total vots as orijinaly kounted. So, the number of voters sysing to expres a preferens must also be noted at the fwt of the relevant kolum. )
Sins C has 12 vots, he or she is deklar'd elekted with 12, minus the
quota of 8 vots, lyving a surplus of 4 vots. With STV, this surplus
vot is not wasted.
Al the most prefer'd kandidat's voters hav an equal rIt to desId
how that kandidat's surplus vot is transfer'd or re-distributed to
sekond preferenses, so al the elekted kandidat's vots ar
transfer'd in proportion to the sIz of C's surplus vot.
The 12 transferabl vots from kandidat C only hav a surplus valu of 4 vots abov the elektiv quota of 8 vots. Therfor, the valu of thys 12 transferabl vots kounts for only 4/12 or 1/3 of a vot ych.
The rul for weiting the kount of the transferabl vots is: transferabl
vots minus quota equals surplus, and the surplus is divided by the
transferabl votes.
( This is Gregory's method or the Senatorial Ruls, naim'd after the yus
of STV in varius Komon-welth senats. )
Say, 4 of the 12 transferabl vots giv ther sekond preferenses to D.
As ych transferabl vot is worth 1/3 of a vot, D gains by 4 tIms
1/3 equals 4/3 vots. Say, 5 of the sekond preferenses go to
A, hw gets an extra 5/3 vots.
Neither D nor A hav yet rych'd the quota of 8 vots to elekt them.
Supos B gets the remaining 3 out of 12 sekond preferenses, worth 3/3
or 1 vot. Sins we'v said B got 7 first preferenses, that transfer valu
of 1 vot is just enuf to elekt B.
B has no surplus vot to transfer, and ther ar no other elekted
kandidats ( hws surpluses wud be transfer'd to desId a kontest for
mor than the thry syts of this exampl ). So, kandidat D, with the
lyst vots, is exkluded. D's 3 1/3 vots ar transfer'd to ther
next preferenses.
Say, E, with 6 vots, piks up 2 1/3 of D's vots, to be elekted
to the third and last syt on the komity. This lyvs A with
6 2/3 plus, at most, 1 of D's vots.
Had E not won by 1/3 of a vot abov the quota, E and A mIt hav
tId for third plas, with 8 vots ych. Then they wud hav to draw
straws or tos a koin for third plas, just as wen tw kandidats for a
singl syt get exaktly haf the vots ych.
| Kandidats | 1st staj | 2nd staj | ktd. | 3rd staj | ktd. |
|---|---|---|---|---|---|
| 1st preferenses: C elekted | Transfer of C's surplus at 4/12 = 1/3 | B elekted | Elimination
of D | E elekted | |
| D | 2 | 4/3 | 3 1/3 | ||
| A | 5 | 5/3 | 6 2/3 | 1 | 7 2/3 |
| E | 6 | 6 | 2 1/3 | 8 1/3 | |
| C | 12 | 8 | 8 | ||
| B | 7 | 3/3 | 8 | 8 | |
| total valid vot: | 32 | 32 | 32 |
For many yers, the Electoral Reform Society wud ad uon vot to the
Droop quota. So, the elektiv quota for uon syt, given 100 voters, wud
be 50 plus uon equals 51 vots. Now they dont bother. Bekaus, if tw
kandidats tI with 50 vots ych, wich kandidat taks the uon syt is
not an elektiv desision. They hav to draw lots.
Similarly, if the thry lyding kandidats in a kontest ych won 33
out of a total 99 vots, the kandidat of thys thry, hw drw the short
straw, wud hav to stand down.
Altho most text-bwks defin the Droop quota as total vots divided by
uon mor than the number of syts, folow'd by a 'plus uon vot', yu
kan forget about the 'plus uon vot'.
For elektorats of mor than 100 voters, it dosnt mater wich
version of the formula yu yus. But for smal numbers of voters, the
'plus uon vot' may prevent enuf kandidats gaining ther ful quota,
to tak al the syts. ( Nor nyd uon ad a smal fraktion of a vot, tho
uon must round-up rather than round-down a quota. )
Richard Lung.