International Journal of Theoretical Physics, Vol. 19, No.

6, 1980
Geomet ri c Mode l for Fundament al Parti cl es
E. P. Bat t ey- Prat t I and T. J. Rac e y
Received October 14, 1979
An attempt is made to show that fundamental particles are manifestations of
the geometry of space-time. This is done by demonstrating the existence of a
purely geometrical model, which we have called spherical rotation, that satisfies
Dirac's equation. The model is developed and illustrated both mathematically
and mechanically. It indicates that the mass of a particle is entirely due to the
spinning of the space-time continuum. Using the model, we can show the
distinction between spin-up and spin-down states and also between particle and
antiparticle states. It satisfies Einstein's criteria for a model that has both wave
and particle properties, and it does so without introducing a singularity into the
continuum.
1. I NT R ODUC T I ON
Th e ol de s t me t h o d of r e c o r d i n g mo t i o n is qui t e s i mp l y t he me t h o d of
l a y i n g d o wn a t r ai l . Th e t r ai l c o n n e c t s t he s t a r t t o t he e n d of t he j o u r n e y .
Ta k e t he e x a mp l e of a d o g t i e d b y a l ong r o p e t o a t r ee. He r u n s t wi c e
a r o u n d hi s k e n n e l a n d t h e n l i es d o wn . Hi s ma s t e r r e t ur ns , sees t he r ope ,
a n d i n s t a n t l y k n o ws t h a t t he d o g ha s c i r c l e d hi s h o u s e t wi ce.
I n a n y t h e o r y of t he c o n t i n u u m t ha t p u r p o r t s t o d e s c r i b e ma t t e r as a
d i s t o r t i o n of s pa c e i n t he ma n n e r f i r s t s u g g e s t e d b y W. K. Cl i f f or d, 2 t he
mo t i o n s of t he ma t t e r mu s t n o t d e s t r o y t he c o n t i n u i t y of t he s pace. An y set
of c u r v i l i n e a r c o o r d i n a t e s u s e d t o ma p t he s p a c e i n t he vi c i ni t y of t he
p a r t i c l e mus t , l i ke t he r o p e a t t a c h e d t o t he crog, p a r t i c i p a t e i n t he a l l o we d
mo t i o n s of t he s pace. Thi s r e q u i r e me n t , we h a v e b e e n a b l e t o s h o w ( see
Ap p e n d i x ) , is e q u i v a l e n t t o t he a s s e r t i on t ha t t he o n l y a l l o wa b l e mo t i o n s
mu s t b e r e p r e s e n t e d b y a s i mpl y c onne c t e d gr oup, n a me l y , t he uni ver s al
cover i ng group of t he Li e g r o u p us e d t o d e s c r i b e a l oc a l p a r t of t he mo t i o n .
1 Present address: RR # 1, Inverary, Ontario KOH 1X0, Canada.
2W. K. Clifford (1956). All those who searched for a unified field theory subscribed to a
geometrical interpretation of matter, including Einstein, Weyl, and Eddington.
437
0020-7748/80/06004)437503.00/0 9 1980 Plenum Publishing Corporation
438 Battey-Pratt and Racey
This model shows that a persistent particle must be described by a compact
group. The simplest compact universal covering group is named SU(2).
The model to be described has SU(2) as its motion group.
2. THE MECHANICAL MODEL: THE SPHERICAL
ROT AT OR
The mechanical model that led to the geometric solution of the Dirac
equation can be constructed roughly by suspending a practice golf ball
from flexible wires that are, in turn, fixed onto a wooden framework. It is
sufficient to choose six wires corresponding to the positive and negative
axes of a three-dimensional coordinate system (Figure 1). The ball can
always be rotated indefinitely without the wires becoming entangled. Each
double revolution returns the system to its original configuration, provided
each wire is kept in the correct relation to all the others.
The simplest workable relation is one that requires all spheres con-
centric with the ball to remain rigid; that is to say, the six points where the
wires intersect an imagined spherical shell surrounding the ball must
Fig. 1. Mechanical model for demonstrating spherical rotation.
Geometric Model for Fupdamental Particles 439
Fig. 2. Configuration of the z axis of the model.
r emai n equi di st ant f r om one anot her , t hough t he shell as a whol e is free to
t ur n a bout the cent er.
Wher eas t he bal l of this model can, in pri nci pl e, be t ur ned a bout a ny
axis st art i ng f r om any initial posi t i on, in pract i ce, because of the rest rai nt s
(flexibility, ext ensi bi l i t y, etc.) of the mat er i al s used, t here is a best way f or
demons t r at i ng t he mot i on:
Posi t i on t he bal l so t hat t he poi nt wher e each wi re is at t ached to t he
bal l faces the poi nt wher e it is at t ached t o t he f r ame. ( Ther e mus t be
enough sl ack in t he wi res to al l ow f or what follows. ) Next , t urn t he bal l
t hr ough 180 ~ about a hor i zont al axis, say t he x axis. The vert i cal wi res t hat
were in t he di r ect i on of t he z axi s now t ake on t he conf i gur at i on of Fi gur e
2. Now r ot at e t he bal l about its vert i cal axis. Her e the conf i gur at i on of t he
z axis wi res ret ai ns its shape a nd r ot at es at hal f t he angul ar vel oci t y of t he
ball.
We can go on r ot at i ng t he bal l i ndefi ni t el y, but it will be f ound t hat
af t er ever y t wo r ot at i ons the syst em ret urns to its ori gi nal conf i gur at i on.
Fi gur e 3 i l l ust rat es t he sequence of conf i gur at i ons of t he model dur i ng one
such compl et e cycle.
The wires of this model can be t hought of as a set of cur vi l i near
coor di nat es used t o descri be t he posi t i ons of poi nt s in a me di um t hat
sur r ounds t he spi nni ng core.
For an al t er nat i ve model , i magi ne a l arge spheri cal cani st er whi ch is
compl et el y filled wi t h a gel at i nous medi um. I magi ne, also, t hat t here is a
smal l magnet i zed steel bal l set i n t he cent er. We will pr es ume a r easonabl e
degree of adhesi on to exist bet ween t he j el l y and t he walls of the cont ai ner ,
and bet ween t he j el l y and t he cent r al ball. By means of ext ernal el ect ro-
magnet s we now i nvert t he steel bal l by hal f a t ur n about t he x axi s of a
coor di nat e f r ame cent er ed on the ball. Because of t he resilience of t he
me di um this can be ef f ect ed wi t hout t eari ng or loss of adhesi on. Agai n,
usi ng t he ext ernal el ect r omagnet i c field, we set t he bal l spi nni ng about t he
440 Battey-Pratt and Racey
~(c) /
( o )
/
- 1/
Fi g. 3. Top: Rot a t i on of t he cor e i s ant i cl ockwi se vi ewed f r o m t he uppe r (posi t i ve) z
di r ect i on. (a) I ni t i al posi t i on. (b) Af t e r a qua r t e r t ur n of t he core. (c) Af t e r hal f a t u r n of t he
cor e. (d) Thr e e qua r t e r s t ur n of t he cor e. (e) Af t er one f ul l t ur n of t he cor e. (f) On e a n d a
qua r t e r cor e t ur ns. (g) On e a n d a ha l f cor e t ur ns . (h) On e a n d t hr ee qua r t e r cor e t ur ns . (i)
Two ful l t ur ns of t he cor e r e t ur ns t he s ys t e m t o t he i ni t i al posi t i on.
Geometric Model for Fundamental Particles 441
z axis. A wave of s t r ai n will r ot a t e i n t he j el l y at hal f t he f r e q u e n c y of t he
steel bal l . Af t er ever y t wo t ur ns of t he steel ball, t he s ys t em will r e t ur n t o
its i ni t i al st at e. Th e j el l y will a c c o mmo d a t e t he mo t i o n wi t hout bei ng t or n
or di s r upt e d i n a n y way. 2
3. T HE MAT HE MAT I C S OF S P HE R I C AL R OT AT I ON
I n t he Ap p e n d i x we ha ve s hown t hat e a c h c onf i gur a t i on of t he
spher i cal r ot a t i on mo d e l c a n b e r e pr e s e nt e d b y a poi nt o n a Euc l i de a n
f our - di me ns i ona l hype r s phe r e ( t he Li e gr oup space). Ta k i n g t he hype r -
s pher e t o be of uni t r adi us, we c a n des cr i be t he t r a ns f or ma t i ons f r o m one
c onf i gur a t i on t o a n o t h e r b y a cl os ed u n i mo d u l a r gr oup. ( Ma ki ng t hi s
choi ce is, i n t he l a ngua ge of q u a n t u m physi cs, equi val ent t o t he p r o c e d u r e
of n o r ma l i z i n g t he wa ve f unct i on. ) I f we cent er t hi s uni t hype r s phe r e at t he
or i gi n of a r e c t a ngul a r Ca r t e s i a n c o o r d i n a t e syst em, a n d let t he ve c t or (1,
0, 0, 0) f r o m t he or i gi n l oc a t e t he poi nt c o r r e s p o n d i n g t o s ome c hos e n
i ni t i al conf i gur at i on, t he n a n y ot her c onf i gur a t i on will be gi ven b y t he
ve c t or ( a , f l , % 6 ) , wher e a2"~"f l 2At " y 2 - J t ' ~ 2 = 1. 3 m r ot a t i on i n t he spher i cal
mo d e c a n be r e pr e s e nt e d b y a n y ope r a t or t hat will t r a n s f o r m vect or s of
t hi s t ype ( or t hei r equi val ent s ) i nt o one anot her . He r e we shal l des cr i be t wo
of t he mo s t usef ul r epr es ent at i ons .
(1) Th e vect or ( a , f l , y , 3 ) c a n be wr i t t en as t he q u a t e mi o n
~ = a + i f l + j T + k 8
w i t h [~b[ 2 ~- t~*r = 0l 2 + 2 _1.../2 _[_ ~ 2 = 1, where ~* i s t h e quat erni oni c c o n -
j u g a t e .
Tr a n s f o r ma t i o n s of t hi s qua t e r ni on i nt o a n y ot her u n i mo d u l a r
q u a t e r n i o n c a n be ef f ect ed by mul t i pl yi ng b y a not he r sui t abl e quat er r t i on.
Thus , u n i mo d u l a r qua t e r ni ons do d u t y f or b o t h t he c onf i gur a t i on ve c t or
a n d t he r ot a t i on oper at or .
2If we replace the canister by a surrounding region of stationary jelly, and the steel ball by a
core of spinning jelly, then, insofar as there is strain neither in the stationary region nor in
the core, and their potential energies are therefore both zero, we may have an example of a
soliton. For a general article on Solitons see Claudio Rebbi (1979), p. 92, with further
bibliography furnished on p. 168. This might be worth pursuing though we, ourselves, have
not done so.
3For example, referring to Figure 3, let (1, 0, 0, 0) stand for configuration (a), (0, 1, 0, 0) for
configuration (c), then 0a) would be (2 -1/2, 2 -1/2, 0, 0), (d) would be ( - 2 -I/2, 2 -1/2, 0, 0),
(e) would be ( - 1, 0, 0, 0) etc.
442 Battey-Pratt and Racey
(2) The quat er ni on a + i/3 +j~, + k8 can be r epr esent ed by t he col umn
vect or
8
y
/ 3
Thi s part i cul ar ar r angement of t he coeffi ci ent s has been chosen t o achieve,
in what follows, conf or mi t y wi t h t he t radi t i onal f or mul at i ons of quant um
physics.
Now because i ( a + i fl + j y + k r ) = - 13 + i a - j 8 + k' t, l eft mul t i pl i ca-
t i on by i has t he same effect on a, fl, y, and 6 as t he mat r i x pr oduct :
0 0 0
0 0 1 0 =
0 - 1 0 0
1 0 0 0
similarly, j and k are, respectively, r epr esent ed by
i 0 - 1 0
0 0 - 1
0 0 0
1 0 0
a nd
O
1 0 0
0 0 0
0 0 - 1
Thus t he full oper at or a + ifl +j~, + k/~ is r epr esent ed by
a - t ~ - y - / 3
8 a /3 - ' t
- / 3 ~
/3 , / - 8 a
Thi s mat r i x can be par t i t i oned as shown. Each of t he quadr ant s is
recogni zabl e as t he mat r i x r epr esent at i on of a compl ex number . Hence we
can wri t e
[ a + i 8 - - ' / + i f l ]
r y + i f l a - - i S
The det er mi nant of ~, is a 2 + 8 2 + ,/2 + fl 2 = 1. Thi s mat ri x is uni t ar y as well
as uni modul ar or "speci al ". Thus, t hi s representation o f spherical rotation
T
A
B
L
E

I
.

C
o
m
p
a
r
i
s
o
n

b
e
t
w
e
e
n

t
h
e

Q
u
a
t
e
r
n
i
o
n
i
c

a
n
d

S
U
(
2
)

R
e
p
r
e
s
e
n
t
a
t
i
o
n
s

o
f

S
p
h
e
r
i
c
a
l

R
o
t
a
t
i
o
n

I

Q
u
a
t
e
r
n
i
o
n
:

S
U
(
2
)

S
p
i
n
o
r

P
h
y
s
i
c
a
l

i
n
t
e
r
p
r
e
t
a
t
i
o
n

o
p
e
r
a
t
o
r

a
n
d

o
p
e
r
a
n
d

o
p
e
r
a
t
o
r

o
p
e
r
a
n
d

O
p
e
r
a
t
o
r

O
p
e
r
a
n
d

i

i

|

i
0

~

0
]

i
l
l

i

[
0

i

,
0
]

[
0
]

L
e
a
v
e
s

t
h
e

m
o
d
e
l

a
s

i
t

i
s

R
o
t
a
t
e
s

t
h
e

c
o
r
e

o
f

t
h
e

m
o
d
e
l

t
h
r
o
u
g
h

1
8
0
~

a
b
o
u
t

t
h
e

x

a
x
i
s

(
f
i
g
h
t

h
a
n
d
e
d
l
y
)

[
0

,
]

E
0
]

~
o
O
.
~
e
o
o
r
o
o
,
.
m
~
o
~

1

O

t
h
r
o
u
g
h

1
8
0
~

a
b
o
u
t

t
h
e
y

a
x
i
s

[
~

_
0

I

[
0
]

R
o
t
a
t
e
s

t
h
e

c
o
r
e

o
f

t
h
e

m
o
d
e
l

i

t
h
r
o
u
g
h

1
8
0
~

a
b
o
u
t

t
h
e

z

a
x
i
s

I
n
i
t
i
a
l

p
o
s
i
t
i
o
n

P
o
s
i
t
i
o
n

a
t
t
a
i
n
e
d

b
y

r
o
t
a
t
i
n
g

c
o
r
e

1
8
0
~

a
b
o
u
t

t
h
e

x

a
x
i
s

f
r
o
m

i
n
i
t
i
a
l

p
o
s
i
t
i
o
n

P
o
s
i
t
i
o
n

a
t
t
a
i
n
e
d

b
y

r
o
t
a
t
i
n
g

t
h
e

c
o
r
e

1
8
0
~

a
b
o
u
t

t
h
e
y

a
x
i
s

f
r
o
m

i
n
i
t
i
a
l

p
o
s
i
t
i
o
n

P
o
s
i
t
i
o
n

a
t
t
a
i
n
e
d

b
y

r
o
t
a
t
i
n
g

c
o
r
e

1
8
0
~

a
b
o
u
t

t
h
e

z

a
x
i
s

f
r
o
m

t
h
e

i
n
i
t
i
a
l

p
o
s
i
t
i
o
n

4 4 4 Battey-Pratt and R a c e y
consists of the special unitary matrices of order 2, f rom which comes the name
SU(2).
The operand form of ~ is
I a + i 8
~' +ifl I
This form is called a spinor, a word coined by P. A. M. Dirac.
Table I shows a comparison between the quaternionic and the SU(2)
representations of spherical rotation together with an interpretation of the
symbols.
4. SPIN
Rotation that is a linear function of time is referred to as spin.
A model of spherical rotation whose initial configuration is given by
the spinor [1] can be rotated into the position [~] by the operator [o 0 ].
This is a rotation of the core of the model by 180 ~ about the z axis. It; i s
represented in the Lie group space by a quarter turn around a great circle
in the four-dimensional hypersphere. An intermediate position is given by
the spinor
[ O S 0 o . i O 0 ] [ e ' ~
0 ]
where 0 is the angular displacement along the great circle. Note that this
represents a rotation of the core of the model by 20 about the z axis. The
rotation that brings the model into this position from the initial configura-
tion is represented by the operator [ co8 e~ ]. Thus, if we introduce the time
parameter, t, then the operator [e~, _ o ] wi l l generate the spin [~'~]
representing core spin of angular ~Telocity 2~o. 4 Similarly,
J
cos ~ot i sin tot ]
i sintot cost a J
generates a spin in which the core of the model has angular velocity 2~0
4The configuration of the z-axis wires of the model (Figure 2) rot at es a t half the angular
velocity of the core, that is, with velocity ~0.
Geometric Model for Fundamental Particles 445
a b o u t t he x axis, a n d
[
cost ot - s i n t o t ]
si n tot cos tot
doe s t he s a me a b o u t t he y axis.
I n t he t r a di t i ona l anal ys i s of spi n, it is us ual t o i mpl y t h a t t he pr oc e s s
of i nver t i ng t he axi s of a s pi nni ng obj e c t is i de nt i c a l t o r ever s i ng t he spin.
Wh e n , however , t he s pi nni ng obj e c t is c o n t i n u o u s l y c o n n e c t e d t o its
s t a t i ona r y e nvi r onme nt , t hi s ceas es t o be t r ue; a n d we mus t ma k e a c a r e f ul
di s t i nc t i on b e t we e n t he i nve r s i on a n d t he r ever s al of spin. Thi s di s t i nc t i on
a f f or ds us i nsi ght i nt o o n e of t h e mo s t f u n d a me n t a l pr oper t i es of e l e me n-
t a r y par t i cl es.
To reverse the z axi s spirt [ eo~ l o n e c a n r e ve r s e t i me t - + - t, or o n e c a n
r e ve r s e t he a n g u l a r vel oci t y o f t he mo t i o n : t o-+--t o. Th e r e ve r s e d s pi n
b e c o me s [e ~ '~' ]. Thi s mo t i o n c a n be g e n e r a t e d b y t he r e ve r s e spi n
o p e r a t o r
["-o '~ 2 , ] f r o m t he i ni t i al c o n f i g u r a t i o n [o ~ ]. We shal l suggest i vel y cal l t hi s
antispin a n d r e f e r t o t he antispin state as o p p o s e d t o t he normal spin state of
s pher i cal r ot at i on.
To invert t he spi n axi s of t he c or e of o u r mo d e l it is ne c e s s a r y t o t u r n
t he s pi nni ng cor e a b o u t o n e of t he axes p e r p e n d i c u l a r t o t he spi n axis, f or
exampl e, t he y axis. Th e o p e r a t o r t h a t will a c hi e ve t hi s is [0 01 ]. The r e -
f or e, t he i nve r t e d s pi n st at e is gi ven b y
[ 0 O 1 ] [ e ~ '~ t] _ ~ [ e iO tl
Now,
o l ] E ei o tl =I o l ] E e=o e=l [ O ]
=I : ] I e = ~
= I e = 0 ] I 0 ]
0 e itot 1
Thus , t he i nve r t e d spi n s t at e is g e n e r a t e d b y t he a c t i on of t h e r ever s e s pi n
o p e r a t o r o n t he i ni t i al c o n f i g u r a t i o n [ o] . Last l y, we h a v e t he i n v e r t e d
ant i s pi n s t at e [ e _ ~ ].
Th e c ont r a s t b e t we e n t he a b o v e f our st at es is i mp o r t a n t f or t he l at er
u n d e r s t a n d i n g of par t i cl e physi cs. Howe ve r , it mu s t be u n d e r s t o o d t h a t
446 Battey-Pratt and Racey
there exists an i nfi ni t y of i nt ermedi at e states of our model . The normal
spin state axis can be ori ent ed bet ween the extremes of z axis spin-up and
spin-down. The i nt ermedi at e inverter is
cos x - si nx ]
s i nx cosx ]
where 2X is the axis angle wi t h respect to the spin-up direction. The
i nt ermedi at e spinor is
I e ito t COS X 1
e i~ot s i n X J
which is generated by
cos tot + i sin tot cos2x
i sin tot sin 2 X
i sin tot sin 2 X ]
cos tot - i sinwt cos 2X
J
from the initial position [~~ The correspondi ng ant i spi n dat a are ob-
L ~mxj"
t ai ned by taking the complex conjugate formulas. There is also a con-
t i nuum of states i nt ermedi at e bet ween normal spin and antispin. Thus,
[ e ] . . . . I r176 ] will be i n t he nor mal
o _0, operat i ng on the l mt l al confi gurat i on styx
spin state for X = 0, and in the ant i spi n state for X = or//2-
5. EQUATI ONS OF MOTI ON OF THE SPHERI CAL SPI N
MODEL
Suppose we have a model of spherical rot at i on whose initial config-
urat i on is given by the spinor
[ a + i ~
Wi t hout loss of generality, we shall suppose spin to be generat ed by the
[ e ,' 0 ] The cent er of the spinning core is stationary, and the
operat or o e-"~ "
p h a s e of the rot at i on is, at each instant, the same t hroughout the space
. l
occupied by the model . It will be useful to assume t hat the st at i onary
framework is large and far away from the core so t hat we can ascribe the
confi gurat i on spinor (which describes the phase of the rotation) t o every
poi nt of the local region.
Geometric Model for Fundamental Particles 447
Ac c o r d i n g t o s ome o t h e r obs e r ve r mo v i n g pa s t o u r s ys t e m wi t h
vel oci t y - g , t he s ys t e m is r e pr e s e nt e d b y t he s pi nor
ito(t -- e'~/c2)
eple # 0
- i ~( t- e-~/~2)
0 ~2 e B
He r e we have us ed t he Lo r e n t z t r a ns f or ma t i on t ' ~ ( t - g 4 / c 2 ) / f l , wher e
f l = ( 1 - v 2 / c 2 ) 1/2 a n d g . f = vxx + Vyy + v z z . Thus , t he mo v i n g obs er ver sees
t he c e nt e r of our mo d e l mo v i n g pa s t h i m wi t h vel oci t y g whi l e t he pha s e of
t he r ot a t i on var i es n o t o n l y wi t h t i me b u t al so f r o m pl ace t o pl ace. I n f act ,
f r o m t he e xpone nt f a c t or t - g . f / c 2, we c a n d e d u c e t ha t he sees e a c h
p a r t i c u l a r phas e of t he mo t i o n s weepi ng f o r wa r d wi t h a vel oci t y of c 2 / [ F [
i n t he di r ect i on of g ( Fi gur e 4). Re gi ons of c o n s t a n t p h a s e f o r m pl anes
pe r pe ndi c ul a r t o t he mo t i o n of t he model .
Th e obs er ver al so r e c kons t ha t t he c onf i gur a t i on r ot at es a b o u t t h e c o r e
wi t h a ngul a r vel oci t y t o ( 1 - v 2 / c 2 ) 1/2. Thi s decr eas e f r o m t he val ue to is a
ma ni f e s t a t i on of t he r el at i vi st i c pr i nci pl e of t i me di l at at i on. On t he ot he r
ha nd, t hi s r ot a t i on c o mb i n e d wi t h t he f i ni t e f o r wa r d mo t i o n of t he pha s e
pr oduc e s a c onf i gur a t i on hel i x whos e pi t ch decr eas es wi t h i ncr eas i ng cor e
phase i n advance o f core \ x x phase lagging b e h i n d core region
I \ L v ~-
wave phase a n d core velocity vectors
(unit time)
Fig. 4. Top: The motion of the wires of Figure 2. The angular velocity of the core is 2~ and
that of the wire configuration is ,o. Bottom: The same with its core center moving to the right
(in the z direction) with velocity ,5. The second figure shows the situation 8~r/~(l - v 2 / c 2 ) t / 2
time units after the first. Note how the wires become helices which rotate contrary to a
corkscrew with respect to the advancing core center. (]'he helix of a corkscrew advances
along a fixed locus.)
448 Bat t ey- Pr at t a nd Ra c e y
velocity (exemplified by a select pair of model wires in Figure 4).
Measured a t t h e p o s i t i o n o f t h e o b s e r v e r , this helical configuration rotates
with angular velocity ~/ ( 1 - v2/ c2) 1/ 2. 5
It should be noticed that the mathematical description of the motion
of our model describes only the configuration phase: it does not indicate
the position of the core center. It follows that a theory entirely based on
spinors and their equations will be essentially incomplete.
Within this incomplete theory, we can formulate the general law of
mot i on for the phase of the configuration surrounding the model by
deriving the differential equation that is independent of 15:
i t ~ ( t - - ~ . ~ / c 2 )
0__~_~= L gale 13
Ot Ot - i ~ ( t - ~.~/c2)
ga2 e $
ico i ~ ( t - ~ . ~ / c 2 )
gal - - ~ e 13
( i t o ) - i ' ~
ga2 - - ~ e
~ I it~(t-- ~'~/c2)
= gal e . 13
- - i ( o ( t - - ~' ~1c 2)
- - ga2 e lJ
0t 2
0 i~
0 t / 3
iro(t--~.~/c2)
gat e P
- - i ~( t - - ~' ~/ c 2)
- - ga2 e P
02 2
=--Sga
Oga = i t O V x
ax c2B
i ~ ( t -- ~'~/c2)
gal e 13
-- ito(t-- 6"?/C2)
- - ga2 e /3
hence
0 2 g a
OX 2 = C - ~ ga
5The di f f er ent be ha vi or of t hes e t wo as pect s of t he a ngul a r vel oci t y i l l ust r at es t he di f f er ence
be t we e n a c ont r a va r i a nt a n d a c ova r i a nt vect or i n Mi nkows ki space, t o o - - v 2 / c 2 ) I/2 is, as
al r eady me nt i one d, r el at ed t o t he t i me c o mp o n e n t of a c ont r a va r i a nt s pace- t i me vect or ;
wher eas , we shal l , wh e n we c o me t o c o mp a r e our mode l wi t h a n e l e me nt a r y part i cl e, see t ha t
w/ ( l - v2/ c2) I/2 is r el at ed to t he e ne r gy of a covar i ant mo me n t u m vect or .
Geometric Model f or Fundamental Particles 449
Likewise,
0 2 ~ 2 2
3y---- ~ ---_ C4fl---~b
Hence, using the notation
we have
THUS,
_ _ _~~ and 0 2~b 2 2
a z 2 - - c 4 3 2 # ,
2,
02t~ "t" 02--'-'-~-~ "4- 0 2 ~
9 = ~ 3),2 3z----7
0) 2 0)21) 2
c , B 2 - - c 4 B 2
1 0 2r = ~222 r
~ 2 t ~ C2 Ot 2
6. FUNDAMENTAL PARTI CLES
There have been various attempts to describe fundamental particles as
forms of the continuum.l But, in conventional thinking, a paradox arises if
particles are considered free to spin. For, if there is indeed a continuum
that at one point is to be ascribed to the particle, and, at another, to the
surrounding void, then the coordinate lines used to map out the whole
space at any given instant would, with the passage of time, become twisted
up and stretched without limit. Alternatively, they would rip, and one part
of the continuum woul d slide past another along a surface of discontinuity.
Now we maintain that the concept of "ripping the vacuum" is intui-
tively absurd, and must be rejected, because it introduces discontinuity
into a theory whose basic assumption is that we should be able to explain
the universe as being continuous. Likewise, an infinite degree of twisting
up of the continuum must also be rejected as unworkable. We therefore
introduce the following postulate:
Postulate. The only allowable persistent motions of a local region of
the continuum with respect to the ambient remainder are such that their
mot i on groups are simply connect ed and compact.
Under these circumstances, the motion in the continuum will be
cyclic, and the system configuration will repeatedly return to each phase of
the cycle (see Appendix).
The simplest possible spinning element in the cont i nuum is one that
rotates in the spherical mode. Imagine, therefore, that the wires of our
model are Gaussian coordinate lines for a reference frame in space. The
450 Battey-Pratt and Racey
ori gi n is in a par t of space t hat is spi nni ng (t he core) rel at i ve to t he
sur r oundi ng part s. The mat hemat i cs t hat we have devel oped so f ar is
t her ef or e appr opr i at e. The conf i gur at i on of t he model becomes t he confi g-
ur at i on of space. Her e we mus t sound a caut i onar y not e. The model was
descr i bed by r ef er ence t o a r ect angul ar coor di nat e syst em i n t he f l at space
in whi ch t he model was pr es umed t o exist. Now we are assumi ng t hat our
spi nni ng conf i gur at i on is t he space itself. Us e of t he model f or mul as
t her ef or e means t hat we shall be descr i bi ng real space by r ef er ence t o a
nonexi st ent backgr ound. Thi s will wor k f or the s ame r eas on t hat enabl ed
Ne wt on to devel op a t heor y of gr avi t at i on on t he as s umpt i on t hat space
was ever ywher e Eucl i dean. But, as was t he case wi t h Ne wt on' s t heory, our
t heor y ma y requi re l at er r ef i nement . 6
Our concl usi on, then, is t hat t he spi nni ng cont i nuum is s ur r ounded by
an undul at i ng, wavel i ke regi on whose phase, @, satisfies the equat i on
1 0 2 t ~ 0 ) 2
v % ' a t 2 =
Thi s l ooks very like t he Kl e i n - Go r d o n equat i on f or an el ement ar y part i cl e.
We shall t herefore ma ke t he i dent i f i cat i on exact . The usual f or mul at i on of
t he Kl e i n - Go r d o n equat i on is
1 0 2x~ m2c 2
V2x~ . . . . X~
C 2 Or2 ~2
wher e m is t he mass of t he part i cl e, h is Pl ank' s const ant di vi ded by 2rr,
and a/ is the wave f unct i on f or t he part i cl e. The wave f unct i on must ,
t herefore, be i dent i fi ed wi t h t he spi nor t hat desi gnat es t he phase of t he
spheri cal spi n in space and t i me. We can now see why t he wave f unct i on
has no physi cal l y meani ngf ul magni t ude, but is usual l y nor mal i zed. The
i dent i fi cat i on requi res t hat we put
m2c 2 r 2
2 C2
whence
I m l = l ' 0 1
6For example, a radially dependent gauge transformation can be added to an exact theory of
particles (that includes a description of the core location) to bring about a contraction in the
volume of the particle's space. This could account for the gravitational field.
Ge o ~ Model f or Fundament al Part i cl es 451
Thus , we de duc e t ha t wh e n we ha ve a s pi nni ng r e gi on of t he c o n t i n u u m, it
i nt er act s wi t h ot he r s pi nni ng r egi ons i n s uch a wa y t h a t we f i nd it
ne c e s s a r y t o i nt r oduc e a me a s ur e f or t he i ner t i a of t he i nt er act i ons . We
ar bi t r ar i l y def i ne t he uni t of mas s . But it wo u l d a p p e a r t ha t t he i ner t i a of
t he s pi nni ng r egi on is s i mpl y a ma ni f e s t a t i on of its a ngul a r vel oci t y; and, if
t hi s is me a s u r e d i n r a di a ns per s econd, t he n t he r at i o of t he a r bi t r a r y ma s s
uni t t o t he spi n f r e q u e n c y is h / c 2, whi c h must , t her ef or e, be a c ons t a nt .
Thi s is t he si gni f i cance of Pl a nk' s c ons t a nt .
Th e as cr i pt i on of ma s s t o t he a ngul a r vel oci t y of t he s pi n is r el at i ve t o
t he obs e r ve r a n d me a s u r e d at hi s l ocat i on. 7 So a par t i cl e, whos e ma s s is m 0
wh e n s t at i onar y, has a ma s s m o / ( 1 - v 2 / c 2 ) 1/2 wh e n i n mo t i o n wi t h
vel oci t y 6 (see pa ge 458). An o t h e r c o n s e q u e n c e of t he i dent i f i cat i on
rnc 2 = hto is t ha t t he c o n f i g u r a t i o n spi n, t ha t is, t he o u t e r mo s t undul a t i ons
of t he syst em, is t he d e B r o g l i e w a v e . Thi s wa ve e mbodi e s a c or e r egi on
t hat , i n t he s t a t i ona r y par t i cl e, spi ns wi t h f r e q u e n c y t o / ~. Thi s is t he
Zi t t e r be we gung t ha t was f i r st s t udi ed b y E. Sc hr 6di nge r (1930). I n t he
mo v i n g par t i cl e, t he Zi t t e r b e we g u n g f r e q u e n c y sl ows t o ( t o / ~r ) ( 1 -
V2/C2) 1/2 (see pa ge 458). 8
To i ncr eas e o u r i nsi ght i nt o t he e l e me nt a r y par t i cl e f o r me d b y t he
s pher i cal l y s pi nni ng ma ni f ol d, we shal l est abl i sh a f i r s t - or der di f f er ent i al
e qua t i on f or t he s pi nor as is d o n e i n rel at i vi st i c q u a n t u m me c ha ni c s . Th e
s t a n d a r d f i r s t - or der o p e r a t o r i n Mi nkows ki s pace is t he q u a d , ["], whos e
s e c ond- or de r der i vat i ve is t he d ' Al e mb e r t i a n [ : f f f i V 2 - O 2 / c 2 0 t 2. T h e
Kl e i n - Go r d o n e q u a t i o n is of t e n wr i t t en i n t he f o r m
( [ ] 2 -- m2c2/~2)xi t ~_ 0
Th e q u a d oper at or , as n o r ma l l y used, is a f our - di me ns i ona l ve c t or ope r a -
t or. We shal l mo d i f y it so t ha t i t b e c o me s a s pi nor oper at or . I n t r adi t i onal
q u a n t u m me c ha ni c s it is usual , at t hi s st age, t o i nt r oduc e a g r o u p of
He r mi t i a n mat r i ces k n o wn as Paul i ' s spi n mat r i ces. We shal l n o t d o t ha t
7We shall use the word "observer" in the way that is usual when discussing di f f erent ref erence
frames. At the level of an elementary particle, the notion of a real observer ceases to have
meaning. We shall retain the word, however, as meaning the person or object (in this
context, the other interacting particle) t o wh o m the relevant calculations apply.
Si n traditional quantum theory, the mathematical formalism that produces the Zitterbewe-
gung term indicates a frequency that increases with the particle's motion. This is because the
formalism is desi gned t o predict the results of act ual meas urement s . A me a s ur e me nt ma de
o n a moving system is made wi t h ref erence to the stationary f r a me wo r k of the measuring
device. It is the projection of the spin of the moving core onto the stationary measuring f rame
that increases in frequency. Most texts o n relativistic quantum theory go through the exercise
of using the established mathematical formalism to calculate a particle's velocity. We give, as
an example, J. McConneU (1960).
452 Battey-Pratt and Racey
he r e as we ha ve n o r e a s o n t o us e Paul i ' s ma t r i c e s a n d no i nt e r pr e t a t i on f or
t hem. We shal l t he r e f or e i n t r o d u c e t he o p e r a t o r
0 8 - i - - i ] ~ z
[: ]-- I 0 i 0 1 0 ]-@Y + [ 0
We h a v e a phys i c a l i nt e r pr e t a t i on f or e a c h t e r m i n t hi s expr es s i on. Th e
ma t r i c e s us e d f o r m a bas i s f o r t he a l ge br a of SU( 2) . Ea c h of t he s p a c e
t e r m ma t r i c e s t ur ns t he de r i ve d s pi nor i n t he t e r m t hr ough a q u a r t e r t ur n
( or t he c or e t hr ough a hal f t u r n ) a b o u t t he a s s oc i a t e d axis. Th e qua -
t e r ni oni c f or m of t he o p e r a t o r is
8 ~ . ~
ic~ )t + i + J ~ y +k- - ~- z
whi c h is s t r ongl y s ugges t i ve of t he us ua l q u a d ope r a t or . Th e q u a t e r n i o n i c
f o r m r e mi nds us t ha t t her e is a c o n j u g a t e o p e r a t o r
1 0] i0 i 01
D * = [ O 1 i O ] ~ x - - [ 0 i 0 l
We n o t e t ha t [ - I *D = [ o 1 o ][:]2, wh e r e E] 2 is t he d ' Al e mb e r t i a n ope r a t or .
Le t us n o w c a l c ul a t e De?, wh e r e e? is t he s pi nor f or s pher i cal s pi n i n
t he ma ni f ol d. To s ave space, si nce t he e xpone nt i a l par t s o f t he s pi nor s
o c c u r so f r equent l y, we shal l i n t r o d u c e t he a b b r e v i a t i o n s
i t o ( t - - 6 . F / c 2 ) - - i t o ( t - - 6 " F / c 2 )
e + = e # , e- - - - - e #
Us i ng t he der i vat i ves f r o m p a g e 457, we s ee t ha t
to e?l e + +
["]e?= - ~ - - e?2e- c2fl k e?le +
1 o [ 0 e + ]
1 + ~ . / c e?,e + ~ - _ ( ~ + i ~ , ) / c
-~ e?~e -
(t)x ~ i.i.i;y )/c e ? le + 1 - f l z / c e ? 2 e -
or
t o !
D ~ = c e ? ( A)
Geomet r i c Model f or Funda me nt a l Par t i cl es 453
where we have defi ned the quant i t y
~ , = f l [ ( l + v J c ) - ( v ~ + i V y ) / C ]
( v ~ - i v y ) / c - ( 1 - v J c ) 'p
Now we have al r eady seen t hat [ : ] * [ ~ = [:]2~ = (0)2/C2)t~. Hence
, t O t ~ 2
[] T ~ = - ~
o r
~o
W e c a n c o m b i n e (A) a n d (B) into o n e equation b y writing
~=
t ~l e +
t ~2e -
( l + v z / C ) ( v ~ + i v y )
f l ePle + cfl
( v x " iVy) (1 - v J c )
eft ~ l e + f l
r -
~b2e -
Then
0 0
1 0
0 1
0 0
o 1
o o~
o ~-~-~
1
+
i 0
0 0
0 0
0 - i
o ]
o o~
- i
0
+
- 1 0
0 0
0 0
0 - 1
Ty
i
0
+
0
0
o o !]
- i 0 ~
0 - i 0z
0 0
0 0
= _ ~ 0 0
c 1 0
0 1
( B )
1 0 ]
0 1
0 0
0 0
454 Battey-Pratt and Racey
The top left quarter part i t i on of the l eft -hand side of the equation, being
the st at ement [[]q~, is equat ed to the lower hal f part i t i on of (t o/ c)~ by
means of the partition-reversing operat or
0 0 1 1
0 0 0
1 0 0
0 1 0
The question nat ural l y arises, is this equivalent to Dirac' s equation?
To see t hat it is, we write {' i n place of 4~, mc / h in place of o~/c,
mul t i pl y by hc, extract the fact or i from the space terms, and rearrange to
f or l n
~
- 1
0
0
- 1
0
+
0
0
- 1
0
0
0
o o ] o
0 0 O~ i
0 1 - ~ x + 0
1 0 0
~176176
1 0 0 0 { ,
0 1 0 ]--~z J l +
0 0 - 1
- i 0
0 0
0 0
0 - i
0
o ok
i 0y
0
0 0 - 1 0 ]
0 0 0 - 1 ] mc2{,
- 1 0 0 0
0 - 1 0 0
O,I,
1 0 ih
= 0 1 - ~
0 0
We can now i nt roduce a similarity t ransformat i on. The mat ri x
A= A- l =
--2 -1/2 0 2 -1/2 0
0 - - 2 - I / z 0 2 -1/2
2-1/ 2 0 2-1/ 2 0
0 2-1/2 0 2-1/ 2
is its own inverse. This gives the following t ransformat i ons:
A I A - I = I,
0
A 0
- 1
0
0 l O 1 [ 1 0 0 o
0 0 - 1 A- l __ 0 1 0 0
0 0 0 0 0 - 1 0
- 1 0 0 0 0 0 - I
Geometric Model for Fundamental Particles 455
( I t wa s t hi s r esul t t ha t d i c t a t e d t he f o r m of A. )
A
0 - 1 0 O]
- 1 0 0 0 J A - I =
0 0 0 1
0 0 1 0
0 0 0 1 ]
0 0 1 0
0 1 0 0 '
1 0 0 0
0 - i 0 0 ]
i 0 0 0 A -
A 0 0 0 i
0 0 - i 0
1=
0 0 0 i |
]
0 0 - i 0
0 i 0 0 '
- i 0 0 0
A
- 1 0 0 0 ]
0 1 0 0 I A -
0 0 1 0
0 0 0 - 1
0 0 1 0
0 0 0 - 1
1 0 0 0
0 - 1 0 0
He n c e o u r s pi nni ng ma n i f o l d e q u a t i o n b e c o me s
+
0 0 1
0 1 0 OxI'
1 0 0 --~-x +
0 0 0
0 0 1 0
0 0 0 - 1
1 0 0 0
0 - 1 0 0
0 0 0 i
0 0 - i 0
0 i 0 0
- - i 0 0 0
O'Y
Oy
- g ; i
1 0 0 0
0 1 0 0
+
0 0 - 1 0
0 0 0 - I
mc2,~
l ~ 1 7 6
0 1 0 ih Ox~
0 0 1 at
0 0 0
whe r e 'I" = A ~' .
Thi s is Di r a c ' s e qua t i on. Thus , we ha ve s h o wn t hat t he ent i t y f o r me d
b y a s pher i cal l y r ot a t i ng d i s t u r b a n c e of t he ma n i f o l d is a Di r a c par t i cl e.
456 Battey-Pratt and Racey
Eval uat i ng xI,, we have
I
--2 1/2 0 2 -1/2 0
"I' = 0 2 - 1 / 2 0 2 - 1 / 2
2-1/ 2 0 2-1/ 2 0
0 2 - 1/2 0 2- 1/2

~1 e+
~b2e -
(l+v,/c)
fl r Cfl
( V x - - i V y ) (1 -- V / C )
cfl 'Pie+ fl
~2 e -
~2 e -
1 l + v ~ / c ) ~ l e + _ ( v x + i v y ~
+(1
~ } ~ b l e -- 2 - - ~ + 2 1 / ~ ~bxe-
( ~ l - ~ + l + v z / c ) ~ ? l e + - - ( V x + i V y ) * 2 e -
2l/2fl 2 1 / 2 f l C
( v x i , , + i 1 1 , . )
} * l e + 2 1 / ~ 21/ 2f l ' 2 e -
For the st at i onary particle, 6 = 0, and
0
x i t = - 2 1 / 2 @ 2 e - i ' a
2 1 /2 ~ p le i~ O t
0
The fact or 21/2 is uni mpor t ant . We st art ed wi t h the normal i zed 2-spinor
and have derived what is called a 4-spinor, q' , which is not
normalized. The normal i zat i on of the 2-spinor was arbitrary, and was
based upon selecting a uni t hypersphere for SU(2). Ha d we chosen a
hypersphere of radi us 2 -1/2, t hen o u r 4-spinor woul d have magni t ude
xI'*xI'-- 17 where ,I,* is the conj ugat e t ransposed row matrix. We coul d t hen
Geometric Model for Fundamental Particles 457
absorb the 2 1 / 2 factors by writing ffl = 2 1 / 2 d ? 1 and t~2 = 21/2~b2 to pr oduce a
normal i zed 4-spinor,
0
_ ~b2e - Rot
~ 1 ei ~
0
i n wh i c h I~1"1~1 + ~2 " ~2 = 1.
Now if we had begun by assumi ng our mani f ol d to be in t he state
[, , e- ] and repeat ed the previous calculations, we woul d have
described by [,2e + 1'
f ound the correspondi ng 4-spinor for the system at rest to be
_ t ~ l e - R o t -
0
0
~ 2 e Rot
All ot her spin directions also satisfy the Di rac equation. For example,
y axis spin represented by the spi nor
s i o , l [ 0 ]
sin ~0t cos ~0t ~'2
(see page 455) corresponds to the renormal i zed Di rac spi nor
( - - ~ 1 + i ~ 2 ) e - i ' ' t
1 ( - - i ~ 1 - - ~ 2 ) e - i ' ' t
2 ( ~ 1 + i ~ 2 ) e i ' t
( - - i t ~l + ~ 2 ) e i ~t
for a particle at rest. The most general Di rac 4-spinor for a particle at rest
is
1
{ ( - - 1 4" n ) ~ l + ( i ) ~ "4- ] ) ~ 2 } e - i ' ~ t
{ ( - / m + 0 ~ , + ( - 1 - , , ) , ~ , } e - " ~ '
{ ( 1 + n)t~l + (i r a + l ) ~ 2 } ei'~
( ( - i m + l )t~l + ( 1 - n)t~2}e i'~t
458 Battey-Pratt and Racey
bei ng deri ved f r om the 2-spi nor
cos o~t + in sin o~t
( m + i l ) s i n o~t
where l, m, n, are t he di rect i on cosines of t he spi n axis.
cos o~t - in sin o~t t ~ 2
7. PARTI CLE STATES
Earl i er (page 455) we sel ect ed a par t i cul ar combi nat i on of initial
conf i gur at i on and spin axis of our rot at i onal mode and l abel ed it t he
"nor mal spin st at e" in or der to cont r ast it wi t h t he "ant i spi n st at e"
obt ai ned by reversi ng t he spin. We also def i ned "i nver si on" of t hose states
f r om t he " up" to t he " down" ori ent at i on. We shall now compar e t hose
l abel ed spin states with t hei r cor r espondi ng 4-spinors.
By compar i ng these results ( Tabl e II) with t hose of st andar d quant um
t heory, we see t hat what we have called the "ant i spi n st at es" pr oduce
solutions t hat are nor mal l y appl i ed to t he el ect ron. Our "nor mal spin
st at es" t herefore cor r espond t o t he posi t ron. The t heory is s y mme t r i c al . We
no l onger need t o post ul at e t he exi st ence of holes in a sea of negat i ve
energy states f or t he posi t r on as is done in st andar d t heory. We assert t hat
all mass and ener gy are cycl i cal di st urbances of t he cont i nuum, and t hat
t hei r measures are pr opor t i onal to t hei r frequenci es. Ther e is, t herefore, no
such thing as negat i ve ener gy any mor e t han t here is negat i ve t emperat ure,
in t he absol ut e sense.
Ther e is no i mmedi at el y obvi ous r eason why t he spi nni ng cont i nuum
must, in pract i ce, be ei t her a part i cl e or an antiparticle. Presumabl y, some
basi c rest ri ct i on forbi ds t he cont i nuum of i nt er medi at e states in free
part i cl es suggested by t he f or mul a at the end of sect i on 4 (page 456). It
woul d appear t hat t he r ot at i on axis is l ocked i nt o t he conf i gur at i on at any
gi ven instant. Of significance is the fact t hat t he ant i spi n states are mi r r or
images of t he cor r espondi ng nor mal spin states (this can be seen by
exami ni ng t he model of spheri cal rot at i on).
We have pr oduced a t heor y of part i cl es and t hei r states that, cont r ar y
to t radi t i onal quant um t heory, is i ndependent of measur ement s made on
t he particles. A particle, as such, is unobser vabl e (see Appendi x, page 456).
Measur ement s can onl y be made by i nt eract i ng with the particle, and t he
onl y t ool s for i nvest i gat i on are ot her regions of di st ort ed cont i nuum (we
have her e made t he pr esumpt i ve leap of assumi ng phot ons t o be undul a-
tions of t he space-t i me structure). Int eract i ons will be gover ned by rules of
wave i nt erference and by what ever geomet r i c conservat i on laws t ur n out t o
be f undament al laws of t he universe.
Geometric Model for Fundamental Particles 459
TABLE II. Comparison of 2-Spinor and 4-Spinor Spin States
Name of state 2-Spinor 4-Spinor
[ ~ 1
o a m . p up I e o ' l 0
e i ~t
o
I ~
e icot
Pmtispin "up" o
0
I ~
o
The negative coefficients of the antispin 4-spinors are of no
significance and can be eliminated by a time shift of ~r/to.
It is cl ear t hat , in or der to f i nd t he posi t i on of t he particle, it is
necessary to l ocat e t he cent er of t he spinning core. Thi s means t hat t he
whol e di st urbance t hat const i t ut es t he part i cl e has t o be confi ned. Thi s
confi ni ng process, whi ch is i mpl ement ed by ot her mani f ol d di st urbances,
consists in so enri chi ng t he har moni c cont ent of t he part i cl e' s de Broglie
wave t hat its mome nt um becomes unpr edi ct abl y altered. Thus, the measur-
ing process is limited by the indeterminancy pri nci pl e--not wi t hst andi ng
which, t he part i cl e does at all t i mes have an exact l ocat i on, and, when
compl et el y free of i nt eract i ons with ot her particles, has a ver y defi ni t e
moment um.
I mpl i ed by t he pri nci pl es t hat have been f or mul at ed in this paper is
t he i dea t hat t here can be mor e compl ex di st urbances t hat do not di srupt
t he cont i nui t y of space. An at om in a given ener gy st at e is pr obabl y an
exampl e of this hi gher - or der compl exi t y. Thi s bei ng so, t he same pri nci pl e
t hat prevent s spheri cal spin f r om exchangi ng r ot at i onal energy wi t h its
surroundi ngs woul d also i nhi bi t t he at om f r om radi at i ng energy. Thi s
allows f or the exi st ence of st eady states of at oms wi t hi n t he st ruct ure of a
classical t heor y of t he cont i nuum, and t hus r ender s Neils Bohr' s ad hoc
hypot hesi s unnecessary.
460 Battey-Pratt and Racey
APPENDI X
A.1 LI E GROUP S AND T HRE E - DI ME NS I ONAL ROT AT I ON 9
I n or der to anal yze t he nat ur e of t hr ee- di mensi onal r ot at i on, we shall
const r uct what is cal l ed t he Lie group space f or t he rot at i on.
Let S be a bal l free to r ot at e a bout its cent er C. Let t he initial posi t i on
of S be r epr esent ed in t he Li e gr oup space by t he poi nt O. I f S is r ot at ed
t hr ough angl e 0 a bout an a xi s / , its new posi t i on will be desi gnat ed by t he
poi nt in t he Lie gr oup space t hat lies 0 units f r om O on t he line t hr ough O
paral l el t o the axis l, and in t he di r ect i on t hat a r i ght - handed cor kscr ew
woul d move if it wer e pl aced al ong l and t ur ned t hr ough t he angl e 0. A
r ot at i on 0 in the reverse di r ect i on about l is r epr esent ed by a poi nt 0 uni t s
f r om O but in t he opposi t e di r ect i on f r om t he pr evi ous poi nt . Because a
r ot at i on of ~r uni t s about l will t ake t he bal l S i nt o exact l y t he s ame
posi t i on as the count er r ot at i on - qr , it follows t hat in t he Li e gr oup space,
we need onl y go as f ar as ~r uni t s f r om O in each of t he di r ect i ons paral l el
t o 1 in or der to end up wi t h a l i near ar r ay of poi nt s r epr esent i ng ever y
possi bl e posi t i on of S t hat l eaves 1 fixed. We can, t herefore, i dent i fy t he
t wo end poi nt s er a nd - ~ r as one a nd the s ame poi nt . Thi s const r uct i on is
r epeat ed for all t he possi bl e or i ent at i ons of t he axi s l.
The Li e gr oup space, t hen, consi st s of a "s pher e" of radi us ~r uni t s
cent er ed on O wi t h di amet r i cal l y opposi t e poi nt s i dent i fi ed. The na me of
this gr oup is t he proper orthogonal group in three dimensions and is denot ed
by t he s ymbol 0 ( 3 ) + . I t is al so somet i mes cal l ed t he t hr ee- di mensi onal
r ot at i on group, but this is a mi s nome r as we shall see in due course.
A Li e gr oup is, f r om a mat hemat i cal st andpoi nt , a fai rl y compl i cat ed
st ruct ure. I t can be t hought of as bei ng a compat i bl e combi nat i on of mor e
basi c st ruct ures. Thus it is, a mongs t ot her things, a topological space, and
f or our pur poses it will be suffi ci ent t o descri be a t opol ogi cal space as t he
vehi cl e f or t hose el ement s of geomet r y t hat are unaf f ect ed b y pl ast i c
d e f o r ma t i o n s - - s u c h as t he cont i nui t y or connectedness of the space: a
pr oper t y t hat is r el at ed to its homotopy type.
I n a t opol ogi cal space, t he pat hs bet ween t wo poi nt s ar e sai d to be
homotopic if t he one pat h can be cont i nuousl y de f or me d i nt o t he ot her. I t
is under s t ood t hat all poi nt s on t he pat h r emai n poi nt s of t he t opol ogi cal
space dur i ng t he t ransi t i on. I f the initial and t er mi nal poi nt s of a pat h
ha ppe n t o coi nci de, it becomes a closed path. By exami ni ng t he cl osed
pat hs t hat begi n a nd end on a par t i cul ar poi nt , we can gai n s ome i dea
about t he t ot al or global st r uct ur e of t he space.
9D. Speiser (1964). This article was the precursor of our theory.
Geometric Model for Fundamental Particles 461
These two
consi dered
one and t he
8~e .
Fig. 5. The rotating object under
consideration.
0 / / " - ~ ;
Fig. 6. The Lie group space of the rotations.
As an illustration, let us t ake t he t wo-di mensi onal space of t he surface
of a t orus (a figure shaped like an anchor ring or doughnut ) . Fi gur e 9 is a
pi ct ur e of it. Emanat i ng f r om t he poi nt P we see t wo paths, p~ andp2, t hat
can cl earl y be def or med i nt o one anot her ; t hat is t o say, t hey are homo-
topic. The col l ect i on of all cl osed pat hs t hr ough P t hat are homot opi c t o
one anot her f or ms a homotopic class, p~ and P2 are t hus in t he same
homot opi c class. The pat hs P3 and P4, on t he ot her hand, cannot be
def or med i nt o one anot her or i nt o pl or p2; t hey are member s of t wo ot her
homot opi c classes. The first class ment i oned, t hat of p~ a nd P2, has t he
uni que f eat ur e of i ncl udi ng t he degener at e pat h, namel y, t he poi nt P itself.
Pat hs like Pl and P2 can be cont r act ed cont i nuousl y unt i l onl y t he poi nt P
remai ns. The poi nt P, t hought of as a degener at e pat h, is cal l ed a nul l pat h.
The set of all homot opi c classes t oget her wi t h t hei r i nt errel at i onshi ps ( t hey
f or m a mat hemat i cal group) charact eri ze t he space and desi gnat e its
homot opy type.
Of special i mpor t ance to Li e gr oup spaces is t he t ype in whi ch t here is
onl y one homot opi c class associ at ed with each poi nt . Thi s class necessari l y
Fig. 7. Two paths which can
be continuously deformed into
one another are homotopic.
Fig. 8. A closed path.
462
Fi g. 9. Pa t hs o n t he s ur f ace of a t or us.
BaO. ey-Pratt and Racey
Fig. 10. Th e Li e gr oup s pace
of 0(3) +.
i ncl udes the null pat h: in ot her words, every cl osed pat h can be cont r act ed
i nt o a poi nt . A t opol ogi cal space of this t ype is descri bed as bei ng simply
connected. If the t opol ogi cal space is also a Li e group, t hen it is cal l ed a
universal covering group, because t here is onl y one simply connect ed Li e
gr oup with a given local st ruct ure, and because all ot her Li e groups havi ng
t he same local st ruct ure, but whi ch are not si mpl y connect ed, are homo-
morphisms of it.
These propert i es and defi ni t i ons will now be exempl i fi ed by r et ur ni ng
to t he discussion of t he or t hogonal gr oup in t hree di mensi ons. We recall
t hat t he group space of 0 ( 3 ) + was a sphere of radi us ~r wi t h di amet ri cal l y
opposi t e points identified.
Consi der t he closed pat hs t hat begi n and end at O i n Fi gure 10. A
pat h Pl t hat remai ns everywhere wi t hi n t he sphere can be cont r act ed i nt o
t he null pat h at O. But what of a pat h t hat crosses t he boundar y of t he
sphere? Let P2 be t he locus of a poi nt t hat starts f r om O and passes out
t hr ough t he surface of t he sphere at poi nt P. Since P is t he same as t he
di amet ri cal l y opposi t e poi nt P' , the cont i nuat i on of this pat h is t r aced by a
poi nt movi ng i nwards f r om t he poi nt P' . If this pat h fi nal l y ret urns to O,
we have a closed pat h. Can this pat h be cont i nuousl y cont r act ed i nt o t he
null pat h? The answer is no. Any at t empt to so cont r act P2 must, at some
stage, r educe the pat h to one t hat is everywhere wi t hi n t he boundar y of t he
sphere, like px. But in or der t o do t hat smoot hl y and cont i nuousl y, t he
poi nt P must sooner or l at er move closer t o P ' unt i l t hey meet . However ,
this is impossible, because P ' is always di amet ri cal l y opposi t e P. Thus, all
t he pat hs t hat cut the boundar y once and t hen r et ur n t o O f or m a
homot opi c class di st i nct f r om t he class t hat includes t he null pat h. Hence,
we have shown t hat 0 ( 3 ) + is not simply connect ed.
Ther e does, however, exist a! Li e gr oup space t hat is l ocal l y t he same
as O( 3) +, but whi ch is si mpl y connect ed. For, consi der t wo spheres of
Geometric Model for Fundamental Particles
463
Fig. 11. The Lie group space of SU(2), which is simply connected, but locally the same as
0(3) +.
radi us ~r cent r ed on O and O' , bot h replicas of 0 ( 3 ) + except t hat a poi nt
on t he boundar y of the first sphere becomes i dent i fi ed wi t h the poi nt on
t he second sphere t hat cor r esponds t o its di amet ri cal l y opposi t e posi t i on
(Fi gure 11). I n this space, we can t r ace a p a t h p f r om O t hat leaves t he first
sphere at a boundar y poi nt P and r eappear s ent eri ng t he second sphere
f r om P' . Af t er crossing t he second sphere, t he pat h leaves at a poi nt Q' ,
and r eappear s in t he first sphere at Q, whence it ret urns t o O. It is now
possible to cont r act t he p a t h p t owar ds O, since we are free t o al l ow P ' to
appr oach Q' (and, si mul t aneousl y, P must move t owards Q). Any cl osed
pat h t hr ough O can, t herefore, be cont r act ed i nt o t he null pat h. Hence, t he
space is simply connect ed. Thi s Li e gr oup space is t he uni versal coveri ng
gr oup of 0 ( 3 ) + , and is called SU(2) f or r easons t hat were given on page
454.
A.2 PHYSI CAL I NTERPRETATI ON (Bolker, 1973)
We recall t hat t he gr oup space 0 ( 3 ) + was a geomet ri cal descr i pt i on
of t he rot at i ons of a ball, S, i n whi ch every posi t i on of S was r epr esent ed
uni quel y by a poi nt in t hat space. The initial posi t i on of S was r epr esent ed
by t he poi nt O.
Let us fi nd t he physi cal meani ng of t raversi ng a small cl osed pat h
t hr ough O. As we l eave O in a gi ven direction, t he ball S t urns about t he
axis parallel t o t hat di rect i on. If we now t ravel across t he Li e gr oup space
keepi ng r oughl y t he same di st ance f r om O, t he ball S will r ot at e about
vari ous axes in such a way t hat it remai ns at r oughl y t he same angul ar
devi at i on f r om t he initial posi t i on. Fi nal l y, as we r et ur n t o O, t he ball
decreases its angul ar devi at i on unt i l it is ri ght back in t he posi t i on f r om
whi ch it started.
464 Battey-Pratt and Racey
m o v e m e n t o f a
p o i n t o n t h e
. s p h e r e
Fig. 12. A small wobble of the sphere S picks out a small path through O in the group space
0(3)+.
How woul d one descri be the mot i on execut ed by S? Clearly, t he
answer is t hat S wobbl ed; t hat is to say, it execut ed t he sort of mot i on t hat
coul d have been i mpar t ed to t he ball if it were hel d fi rml y i n the hand, t he
wrist and el bow gi ven a few flexes and twists, and t hen rel axed back i nt o
t he original posi t i on.
We can now give a physi cal i nt er pr et at i on of homot opy. A closed
pat h on O t hat is j ust a shade di spl aced f r om t he previ ous pat h represent s
a wobbl e of S in whi ch t he mot i ons occur ar ound axes t hat are slightly
di spl aced f r om the pr evi ous axes, and to an ext ent t hat is a shade di fferent .
The t ransi t i on f r om one pat h to t he ot her, t hen, consists of ei t her ex-
aggerat i ng or di mi ni shi ng t he mot i on. Thus, t he sequence of pat hs in t he
Li e gr oup space const i t ut i ng t he smoot h t ransi t i on bet ween t wo homot opi c
pat hs requi res t hat t he mot i ons of t he ball S under l yi ng t he one pat h be
capabl e of smoot h and gr adual exaggerat i on unt i l t hey coi nci de wi t h t he
mot i ons under l yi ng t he fi nal pat h. All t he pat hs of t he homot opi c class
t hat includes t he null pat h are descri babl e as wobbl es of a great er or lesser
ext ent ; t hat is, t hey can be demonst r at ed with S hel d fi rml y in t he hand.
I n 0 ( 3 ) + , a closed pat h t hat leaves O, t hat moves st rai ght al ong a
radi us unt i l it leaves t he surface of t he space, t hat r eappear s at t he
di amet ri cal l y opposi t e poi nt , and t hen moves straight back to O corre-
sponds t o one compl et e r evol ut i on of S ar ound an axis paral l el t o t he pat h.
Thi s pat h is not in t he same homot opi c class as t he null pat h. I n SU(2),
however, t he Li e gr oup space of 0 ( 3 ) + is r epr esent ed twice. A cl osed
straight pat h t hat leaves O al ong a radi us crosses t he second space via t he
di amet er t hr ough O' bef or e r eappear i ng in the first space and r et ur ni ng t o
O. The ball S, in this case, makes two compl et e t urns ar ound an axis
paral l el t o t he pat h. But SU(2) is si mpl y connect ed; t herefore, t he pat h
Geometric Model for Fundamental Particles 465
Fig. 13. A closed path in SU(2) consisting of two parallel diameters through O and O'
represents a double rotation of S.
described is homot opi c to, and, so, cont ract i bl e i nt o the null pat h. This
means t hat the mot i on of S (i.e., a compl et e doubl e r ot at i on ar ound an
axis) can be arrived at by a smoot h exaggerat i on of a wobble, which, in
turn, means t hat it is possible t o grasp a ball in the hand, and, by flexing
of wrist and elbow j oi nt s, etc., to execute a compl et e doubl e (but not
single) revol ut i on of the ball about a fixed axis, and t hereby end up in the
same st ance as one began.
This is, in fact, the basis of an ol d part y trick in which t he j oker twirls
a bowl of soup ar ound its vertical axis so as not to spill a drop. The trick
requires raising and lowering the bowl somewhat duri ng t he gyrat i on to
accommodat e the awkwar d j oi nt i ng of the arm; but it is, nevertheless, an
illustration of the above t heory. The bowl makes two full t urns in execut-
ing the complete gyrat i on.
We can replace the huma n arm and the soup bowl by a l ong coiled
spring and the ball, S. The t op of the spring can be cl amped t o a st at i onar y
f r amewor k in i mi t at i on of t he shoul der connect i on, and the bot t om end
S
Fig. 14. Coiled spring simulating a human arm.
466 Battey-Pratt and Racey
i I 1 1 ~ - ~ .
I ' < v . L ; I J
a . .
hX /
~ /// I
/_.,
Fig. 15. Each half turn of S forms a twist in the two tracers (dotted lines), which can only be
undone by the displacement shown (solid lines).
c a n be b e n t a r o u n d a n d b r o u g h t u p u n d e r n e a t h t he ba l l i n i mi t a t i o n of t he
h a n d ( Fi g u r e 14). Th e u n i f o r m f l e xi bi l i t y of t he s p r i n g a l l ows t he b a l l t o
s pi n wi t h o u t t he a wk wa r d v e r t i c a l mo v e me n t t h a t oc c ur s i n t he s o u p b o wl
t r i ck.
Th e mo t i o n of t he ba l l a n d s p r i n g c a n b e u n d e r s t o o d i n d e t a i l b y
c o n s i d e r i n g t wo l i nes d r a wn o p p o s i t e one a n o t h e r d o wn t he l e ngt h of t he
u n s t r a i n e d s pr i ng. Th e s p r i n g i s f o r c e d t o b e n d i n a s s u mi n g t he a b o v e
c o n f i g u r a t i o n , b u t i t wi l l s wi ng i nt o t he p l a n e t h a t mi n i mi z e s i t s t or s i on.
E x a mi n a t i o n of t he d i a g r a ms of Fi g u r e 15 s hows t h a t t he c o n f i g u r a t i o n of
t he s p r i n g wi l l r o t a t e at ha l f t he a n g u l a r ve l oc i t y o f t he bal l . 1~
No w, b e c a u s e of t he s h a p e of t hi s s ys t em, we c a n a d d a n o t h e r s p r i n g
s y mme t r i c a l l y o p p o s i t e t o t he o n e a bove . I t , t oo, mu s t a d v a n c e a t ha l f t he
a n g u l a r ve l oc i t y of t he b a l l a n d wi l l t h e r e b y ma i n t a i n i t s r e l a t i o n s h i p wi t h
t he f i r s t s p r i n g ( Fi g u r e 16). Th e q u e s t i o n n a t u r a l l y ar i s es : h o w c o mp l i c a t e d
a s y s t e m of s pr i ngs c a n p a r t i c i p a t e i n t he mo t i o n ? To a n s we r t hi s, we go
b a c k t o f i r st pr i nc i pl e s .
Co n s i d e r a b a l l t o whi c h i s a t t a c h e d a l a r ge i nde f i ni t e n u mb e r of l o n g
n a r r o w s pr i ngs r u n n i n g r a d i a l l y o u t wa r d s a n d t i e d a t t he i r o u t e r e n d s t o a
f i xe d f r a me wo r k ( Fi g u r e 17). Si nc e t he s pr i ngs a r e f l exi bl e, i t i s p o s s i b l e f or
t he ba l l t o e xe c ut e s ma l l wo b b l e s a b o u t i t s c e n t e r wi t h o u t c a u s i n g a n y
10Mentioned in Bolker (1973) are both Dirac's spanner--a demonstration of spherical
rotation using strings attached to an asymmetric wrench--and, also, D. A. Adams's
patented device for transmitting electricity to a rotating turntable without the use of slip
rings. The Adams device is identical in its symmetry to our incipient model. A full and
well-illustrated report on it appeared in Adams (1975). Just for the record, we would like to
point out that our model depicted in Figure 1 was constructed and demonstrated in the
Mathematics Department at Queen's University, Ontario, Canada in the fall of 1966. E. P.
Battey-Pratt would like to acknowledge his indebtedness to Prof. A. J. Coleman for
inspiration and to Hans Kummer who, in conversation, helped clarify the role of universal
covering groups in the theory.
Geometric Model f or Fundnment al Particles 467
/
Fig. 16. A symmetric pai r of springs. Fig. 17. Rays of springs emanat i ng
f r om a ball.
ent angl ement of t he springs, But we have seen t hat a doubl e r ot at i on of t he
ball can be devel oped by t he cont i nuous exaggerat i on of a small wobbl e.
Now, t he t ransi t i on f r om a mot i on t hat does not cause t he springs t o tangle
and knot up t o one t hat does is necessari l y abr upt and di scont i nuous, and,
t herefore; cannot occur dur i ng our cont i nuous devel opment of t he wobbl e
i nt o a doubl e rot at i on.
Thus, we see t hat t he springs i n Fi gure 16 are but t wo of an infinite
number of "r ays" t hat can par t i ci pat e in t he mot i on of t he ball wi t hout
knot t i ng up. To f i nd t he di sposi t i on of t he ot her rays, we simply not e t hat ,
st art i ng f r om t he model of Fi gur e 17 we must give t he bal l a hal f t ur n
about a hor i zont al axis in or der t o put t he vert i cal springs i nt o t he
conf i gur at i on of Fi gur e 16. The ball may t hen be r ot at ed about its vert i cal
axis. Thi s is t he expl anat i on f or t he val i di t y of t he model s of Sect i on 2 of
t he paper (page 448).
2:
Fig. 18. Springs i n t he pl ane perpendi cul ar Fig. 19. As the ball is rot at ed about t he z
t o t he initial hal f t ur n about t he x axis. axis, the axis of t he i ni t i a l ha l f t urn also
rot at es,
468 Battey-Pratt and Racey
Now let us go back and fol l ow t hr ough t he same argument s f or t he
case of t wo-di mensi onal rot at i on. We shall t hen pr oduce a geomet ri cal
anal ogy t hat will i l l umi nat e t he rel at i onshi p bet ween t he above analysis
and t he convent i onal view of t hr ee- di mensi onal rot at i on.
A.3 TWO- DI MENS I ONAL ROTATI ON
The Li e gr oup space f or t he t wo-di mensi onal r ot at i on of a fl at obj ect
about some poi nt i n the pl ane is ver y simple. For conveni ence, let t he fl at
obj ect be a ci rcul ar disk D r ot at i ng about its cent er. Let t he initial posi t i on
of t he disk be r epr esent ed by t he poi nt O in t he Li e gr oup space. A
count ercl ockwi se r ot at i on of D t hr ough an angl e 0 is r epr esent ed by a
poi nt 0 uni t s to t he fi ght of O in t he gr oup space. Similarly, a clockwise
rot at i on, - 0, of D is r epr esent ed by a poi nt 0 uni t s t o t he l eft of O. Si nce a
r ot at i on ~r brings D to t he same posi t i on as t he count er r ot at i on - , r , we
can i dent i f y these t wo poi nt s. Thus, t he Li e gr oup space is a line of l engt h
2~r uni t s with end poi nt s i dent i fi ed. Thi s may be t hought of as bei ng t he
ci r cumf er ence of a circle with O and the dual poi nt ~r, - ~r di amet ri cal l y
opposi t e. The name of this Li e gr oup is the proper orthogonal group in two
Fig. 20. The rotating disk.
T h e s e two p o i n t s
to be identified.
Fi g. 21. The Lie group space 0(2)+.
Geometric Model for Fundamental Particles 469
Fig. 22. T h e group space R derived f r o m 0 ( 2 ) + .
dimensions, or 0( 2) +. It is often referred to, erroneously, as the two-
dimensional rotation group.
Since the Lie group space is one-dimensional, the only type of path
that is homotopic to the null path through O is one that leaves O, goes to
some other point, and then returns to O again by tracking back along its
outward path. 0(2)+ is, therefore, not simply connected since it is possi-
ble, by completing one or more circuits of the space, to return to O without
back tracking.
In order to construct the universal covering group of 0(2)+ (the
locally isomorphic but simply connected group space), we duplicate the
space as we did for 0(3)+, and make the end point of the original space
conterminal with the beginning point of the repeated segment rather than
terminating with the other end of itself. We see, however, that we still
cannot rejoin the other end point of the repeated space to the open end of
the original because to do so would produce a circle, which is not simply
connected. We have, therefore, to repeat the segment once more. Again,
we come to an end that cannot be joined back to the first segment without
creating a circular, noncontractible path. Thus, the repetitions of the
segment never end, and the Lie group space of 0(2)+ must be extended
indefinitely in repetitions of itself in either direction. The universal cover-
ing group of 0(2)+ is, therefore, an infinite line. Because every point on
an infinite line can be represented by a real number, the symbol for this
group is R. It is the same as the Euclidean one-dimensional continuum.
A closed path in R is effectively just a line segment that is traversed
both ways, both out and back again to the starting point. Corresponding to
such a path, the disk D turns by the specified amount, and then turns back
again by the same amount to its starting position. That is to say, if one end
of a string were affixed to the rim of D and the other end to a stationary
framework, then a motion that corresponds to a closed path in R would be
one in which, at the completion of the motion, the string would be back in
its original configuration. For, if D makes several full turns in one
direction, then the string (which, remember, is confined to the plane in this
case) becomes wrapped around D that many times. But D then has to turn
back again by the same number of turns in order to complete the motion
while the string unwinds itself again and so returns to its starting config-
uration.
470 Battey-Pratt and Racey
Physi cal l y speaking, j ust as t he mot i ons of t he ball S cor r espondi ng t o
ever y closed pat h i n SU(2) can be devel oped by t he cont i nuous exaggera-
t i ons of a slight wobbl e, so can t he mot i ons of t he disk D cor r espondi ng t o
closed pat hs in R be devel oped by t he cont i nuous exaggerat i ons of a small
osci l l at i on of D.
A.4 CYLI NDRI CAL AND SPHERI CAL ROTATI ON
Hi t hert o, mat hemat i ci ans have r egar ded 0 ( 2 ) + and 0 ( 3 ) + as rot a-
t i on groups. The concept of " r ot at i on" implies mot i on so t hat we are as
much concer ned wi t h t he dynami c process of a body getting f r om its initial
to its final posi t i on as we are wi t h t he posi t i ons themselves. At t he
begi nni ng of this paper, we descr i bed how the movement s made by a dog
tied by a long l ead coul d be t r aced by the fact t hat t he r ope was wr apped
ar ound his kennel. Thi s is a ver y ol d idea t hat deserves to be rai sed t o a
pri nci pl e: the final configuration of a line connecting a moving point to an
observer's stationary framework represents the motion of that point.
Because, superficially, we di scern t he universe as bei ng di vi ded i nt o
mat t er and empt y space, and because we habi t ual l y t hi nk of t he l at t er as
bei ng t he very epi t ome of not hi ngness, we are accust omed t o t hi nki ng of
mot i on as bei ng somet hi ng abs ol ut e- - s omet hi ng uni mpeded by i nt errel a-
tionships. In t he past, mat hemat i ci ans have descri bed mot i on si mpl y by
giving t he begi nni ng and t he end of a j our ney. However , in this paper, we
mai nt ai n t hat mot i on is relative, and t hat t here is, at all times, a t opol ogi -
cal connect i on bet ween rel at i vel y movi ng objects. I n t wo-di mensi onal
rot at i on, t herefore, t he conf i gur at i on of t he string (the dog' s leash) char-
acterizes the r ot at i on of D. In fact, if t he string is pl ayed out f r om a fi xed
poi nt next to t he r i m of D, if it is kept taut, and if D is of uni t radius, t hen
t he l engt h of t he string becomes the r adi an measure of the r ot at i on of D.
Fur t her mor e, it is positive or negat i ve accor di ng to whet her it is wound
ont o D in a count ercl ockwi se or clockwise direction. It is also the measur e
of t he di st ance f r om O t o t he r el evant poi nt (designating t he rot at i on) in
t he gr oup space R. Consequent l y, we can see t hat t he r ot at i on group in t wo
di mensi ons is R not 0( 2) +.
f i x e d p o i n t
f
Fig. 23. String wi ndi ng around a di sc rotating i n the plane.
Geometric Model for Fundamental Particles 471
Similarly, we have descr i bed t hr ee- di mensi onal r ot at i on by at t achi ng
one end of a st ri ng t o a r ot at i ng obj ect and t he ot her end t o a st at i onar y
f r amewor k. I t is t he conf i gur at i on of t he st ri ng t hat char act er i zes t he
r ot at i on. We have, in this paper , er ect ed a model in whi ch all t he possi bl e
conf i gur at i ons of a st ri ng ar e r epr esent ed by poi nt s in t he gr oup space
SU(2). Thus, the rotation group in three dimensions is SU(2), not 0 ( 3 ) + .
And, f ur t her mor e, two f ul l turns of this model are equivalent to not having
t urned at all !
Thi s seems t o conf ound our ver y experi ence. To expl ai n the anomal y,
we shall dr aw upon an anal ogous exampl e in geomet r y. Suppose we
consi der a circle in a pl ane, a nd t hen make a list of s ome of its sal i ent
propert i es. For exampl e:
(1) I t is the l ocus of a poi nt t hat moves so as to be a cons t ant
di st ance f r om a fi xed poi nt cal l ed t he cent er.
(2) It s equat i on in t he s t andar d Car t esi an coor di nat e syst em is
x 2 +y2 = r 2
No w if we ask f or t he t hr ee- di mensi onal anal og of t he circle, t wo possi bi l i -
ties arise: ei t her we ma y dr aw t he circle out in t he t hi rd di r ect i on
per pendi cul ar t o t he pl ane so as to f or m a fi gure in whi ch we can ma p
every poi nt i nt o a line. I n this case t he above pr oper t i es t r ans f or m i nt o t he
fol l owi ng:
(1' ) I t is t he l ocus of a line t hat moves so as t o be a const ant di st ance
f r om a fi xed line caned t he axis.
(2' ) It s equat i on in t he s t andar d Car t esi an coor di nat e syst em is
X2 +y2 = r 2
Or, we ma y al l ow t he t hi rd di mens i on to par t i ci pat e symmet r i cal l y wi t h t he
initial t wo and end up wi t h a fi gure whose pr oper t i es are as follows:
(1") I t is t he l ocus of a poi nt t hat moves so as to be a cons t ant
di st ance f r om a fi xed poi nt cal l ed t he cent er.
(2") It s equat i on in t he s t andar d Car t esi an coor di nat e syst em is
x 2 +y 2 + z 2 = r 2
These figures are, of course, respect i vel y, t he cyl i nder and t he sphere.
Similarly, we have t wo- di mensi onal r ot at i on wi t h its gr oup R and
punct i f or m cent er. Ther e are, agai n, t wo pr i nci pl e ext ensi ons of t hi s
dynami c fi gure i nt o t hree di mensi ons. The fi rst is what we shall cal l
cylindrical rotation. Her e, t he cent er of r ot at i on i n t he pl ane is dr awn out
i nt o an axis of r ot at i on, and t he mat hemat i cal descr i pt i on of it is agai n t he
4 7 2 Ba t t e y - Pr a t t a nd R a c e y
gr oup space R. Thus, angl e agai n becomes t he appr opr i at e measur e f or this
t ype of rot at i on. It is, f ur t her mor e, t he onl y t ype t hat has ever been
consi der ed in t he cor pus of classical mechani cs.
The second t ype is what we shall call spherical rotation, whose
mat hemat i cal descr i pt i on is t he gr oup space SU(2). I n it, t he t hree direc-
t i ons of space par t i ci pat e mor e symmet ri cal l y i n t he mot i on) 1 I t has not
hi t her t o been recogni zed or ful l y under s t ood for t he r eason t hat har dl y any
exampl es of it occur in t he wor l d of our di rect physi cal experi ences. All of
t he r ot at i ng things we nor mal l y s ee- - s pi nni ng tops, mot or shafts, wheels,
et c. - - exhi bi t r ot at i on of t he cyl i ndri cal t ype. To expl ai n why this is, we
reemphasi ze t hat t he string met hod of classifying r ot at i on i ncor por at es a
descri pt i on of t he connect i on bet ween t he obj ect and a st at i onar y f r ame-
work.
Wi t h cyl i ndri cal rot at i on, its gr oup R is infinite or noncompact (to use
t opol ogi cal t ermi nol ogy). If an obj ect rot at es in this mode, t he poi nt
represent i ng t he string conf i gur at i on moves st eadi l y out al ong R in one
di r ect i on to become f ur t her r emoved f r om its st art i ng posi t i on. Physi cal l y
speaking, t he string becomes mor e and mor e wound ont o t he r ot at i ng body
(or twisted up of f t he end of it), and so a connect i ng medi um bet ween t he
body and a st at i onar y f r amewor k woul d have t o possess infinite powers of
ext ensi on. In f act such a medi um does not exist, and t he onl y t wo
al t ernat i ves f or avoi di ng this conf usi on charact eri ze cyl i ndri cal rot at i on.
The first is t hat t he connect i ng medi um shears and so devel ops a surface of
di scont i nui t y. Such a surface, cal l ed a bear i ng surface, is well known i n
mechani cal exampl es wher e t here has to be a suppor t i ng connect i on
bet ween t he rotatirrg body and its st at i onar y r ef er ence frame. The ot her
possibility is t hat t he connect i ng medi um holds, but t hat t he st at i onar y
f r amewor k yields to t ake up t he mot i on of the rot at i on. I n this case, t he
r ot at i on is t r ansf er r ed t o t he connect ed body. Ther e are, t hen, t hese t wo
aspect s of cyl i ndri cal r ot at i on; namel y, t he exi st ence of bear i ng surfaces,
and (or) t he t ransmi ssi on of t he r ot at i on t o or f r om the movi ng body.
On t he ot her hand, with spheri cal rot at i on, t he group space SU(2) is
compact, and t he pat h represent i ng a st eady r ot at i on about a fi xed axis is
closed and finite, so t hat t he string passes t hr ough a cycllw conf i gur at i on t o
keep r et ur ni ng to its initial posi t i on. The i nt er connect i ng medi um in this
case does not shear, nor does t he st at i onar y f r amewor k have t o yield. Thus,
nThe concept of an "axis" of rotation is now severely localized. It only e x t e n d s o v e r t h e c o r e
region, that is, the ball S in our above-quoted models. This ball could, theoretically, be
c o n t r a c t e d to zero radius, leaving only a rotating point a n d r e undulating medium. This
would be the strict model of spherical rotation in which the analogy with the geometric
s p h e r e w o u l d b e e x a c t .
Geometric Model for Fupdamental Particles 473
spherical rotation is characterized by the following: namely, the rotating
body does not transmit or receive rotational energy, but is surrounded by a
medium that undergoes a cyclical "wave" motion. This makes the notion
of spherical rotation elusive to our senses, because, in the world of
macrophysics, motion has generally to be maintained against dissipative
forces like friction. But spherical rotation cannot receive energy from a
propulsive source. In the example given near the beginning of this paper of
a steel ball embedded in jelly, it would be impossible to sustain the motion
mechanically, say by a driveshaft or the like, because such a connection
would necessarily act in the cylindrical mode. An effective driveshaft
would clash with the undulating motions of the jelly and rip it to pieces.
For this reason, we suggested that the ball be magnetic and be propelled
by magnetic forces. A mechanical model of spherical rotation may still be
propelled by the overriding magnetic forces acting in the cylindrical mode.
(Note, however, that an electromagnetic example of spherical rotation
could not exchange electromagnetic energy.) There are other ways to build
visible mechanical models of spherical rotation, but to run them requires
some sophisticated device that is essentially a bit of trickery as far as the
purity of the model is concerned (see footnote 10).
5A. MATHEMATICAL REPRESENTATION OF SPHERICAL
ROTATION
In Section A.1, we stated that the Lie group space SU(2) consisted of
two spheres and their interiors. If these two spheres were superimposed,
their surface points would be identified in diametrically opposite pairs. By
examining an analogous situation in two dimensions, we shall see how to
construct a more integrated model of this space.
Consider two disks with their perimeter points identified in what
would be diametrically opposite pairs if the two disks were superimposed
(Figure 24). If these two disks were embedded in a three-dimensional
Euclidean space and deformed into two hemispherical surfaces, they would
fit together with identified point-pairs coinciding so as to form a sphere.
Thus, the whole space consisting of the two disks becomes equivalent to a
complete spherical surface.
Likewise, the two spherical surfaces of SU(2) can be embedded in
four-dimensional Euclidean space and deformed into two four-dimensional
hyperhemispheres. They can then be brought together over a spherical
surface of contact in such a way that all identified point pairs coincide.
The result is the bounding volume of a four-dimensional hypersphere. The
474 Battey-Pratt and Racey
2~
Fig. 24. Two disks, in which the diametrically opposite circumferential points on alternate
disks are identified, are together equivalent to the surface of a sphere. The centers C, C' end
up diametrically opposite on the sphere.
cent ers O and O' of t he ori gi nal t wo component s of SU(2) become
di amet ri cal l y opposi t e points.
How large is this hyper spher e? We recall t hat in t he space of SU(2),
each sphere had a radi us of ~r uni t s cor r espondi ng t o t he r adi an measur e of
hal f a t ur n of t he physi cal body bei ng rot at ed. Thus, a compl et e ci rcui t of
SU(2) cor r esponds t o a doubl e t ur n of t he ball and has a length of 4~r
units. I n a hyperspheri cal model of SU(2), this circuit cor r esponds t o one
of t he great circles f or med by t he i nt ersect i on of t he hyper spher e wi t h a
pl ane t hr ough t he cent er. It follows t hat t he radi us of t he hyper spher e is 2
units. I n switching our at t ent i on f r om the mot i on of t he ball to t hat of t he
string conf i gur at i on (of Fi gur e 16), we not e t hat it r ot at es at hal f t he
f r equency of t he ball. Ther ef or e, if we use t he measur e of its phase angl e t o
i nduce a measure on t he hyper spher e, the l at t er will have a radi us of one
unit. Of key i mpor t ance is the f act t hat each poi nt in t he boundi ng vol ume
of t he hyper spher e represent s, uni quel y, a par t i cul ar conf i gur at i on of t he
spheri cal rot at i on. We lose not hi ng by decl ari ng t hat t he radius of t he
hyper spher e can be chosen t o be any length.
SUMMARY
Just as the circle in t he pl ane relates to bot h t he cyl i nder and t he
sphere, so, too, does r ot at i on i n t he pl ane rel at e to t he cyl i ndri cal and
spheri cal modes of t hr ee- di mensi onal rot at i on. Classical mechani cs has,
hi t hert o, been defi ci ent in t hat it has onl y r ecogni zed t he cyl i ndri cal mode.
Spherical r ot at i on is the simplest mode in whi ch one par t of space can spin
i n rel at i on to anot her wi t hout di srupt i ng its cont i nui t y. A movi ng vor t ex
spi nni ng in the spheri cal mode satisfies Di rac' s equat i on f or an el ement ar y
particle. Hence, mass and ener gy can be expl ai ned as bei ng mani fest at i ons
of t he r ot at i on of space. Our geomet r i c model demonst r at es t he di f f er ence
Geometri c Model for Fundament al Particles 475
not only between spin-up and spin-down states, but also between the particle
and its antiparticle.
The spherical rotation of space has a spinning center that can be
identified with the position of the particle. As such, hidden variables exist;
that is to say, there always exist an exact particle location and, generally, an
exact momentum. It is an intrinsic feature of spherical rotation that the spin
cannot be transmitted to or from the core, and, therefore, the position of
the particle can be approximated only by bracketing the whole region of
the disturbance--a procedure that also renders the momentum inexact. In
this way do measurements become circumscribed by the indeterminacy
principle.
REFERENCES
Adams, D. A. (1975). Patented devi ce reported i n "The Amat eur Sci enti st, " Scientific
American, Dec., !20.
Bolker, E. D. (1973). " The Spinor Spanner, " American Mathematical Monthly, Nov. , 977.
Cl i f f ord, W. K. (1956). " On the Space Theory o f Matter, " The World of Mathematics, Vol. 1,
Ed. J. R. Newman, (Simon & Schuster, Ne w York), p. 568.
McConnel l , J. (1960). Quantum Particle Dynamics, 2nd Edn. ( Nort h- Hol l and, Ne w York),
pp 116 ff.
Rebbi , C. (1979). " Sol i t ons , " Scientific American, Feb. , 92, t68.
Schrodinger, E. (1930). Preuss. Akad. Wiss., Phys-math. KI. Sitzber., 418.
Speiser, D. (1964). "Theor y o f Compact Lie Groups and Some Appl i cat i ons to El ement ary
Particle Physi cs, " Group Theoretical Concepts and Methods in Elementary Particle
Physics, Ed. F. Gursey, (Gordon & Breach, Ne w York).