Basic structured grid generation with an

introduction to unstructured grid generation 1st

Edition M Farrashkhalvat pdf download

https://ebookname.com/product/basic-structured-grid-generation-
with-an-introduction-to-unstructured-grid-generation-1st-edition-
m-farrashkhalvat/

Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Introduction to Power Generation Technologies 1st

Edition Andreas Poullikkas

https://ebookname.com/product/introduction-to-power-generation-
technologies-1st-edition-andreas-poullikkas/

Smart Grid Security Innovative Solutions for a

Modernized Grid 1st Edition Florian Skopik

https://ebookname.com/product/smart-grid-security-innovative-
solutions-for-a-modernized-grid-1st-edition-florian-skopik/

Grid Resource Management Toward Virtual and Services

Compliant Grid Computing 1st Edition Frederic Magoules

https://ebookname.com/product/grid-resource-management-toward-
virtual-and-services-compliant-grid-computing-1st-edition-
frederic-magoules/

Passing It On A Memoir 2007 3rd printing Edition Yuri

Nakahara Kochiyama

https://ebookname.com/product/passing-it-on-a-memoir-2007-3rd-
printing-edition-yuri-nakahara-kochiyama/
Designing Online Learning with Flash 1st Edition David
Richard Moore

https://ebookname.com/product/designing-online-learning-with-
flash-1st-edition-david-richard-moore/

Three Cities After Hitler Redemptive Reconstruction

Across Cold War Borders 1st Edition Demshuk

https://ebookname.com/product/three-cities-after-hitler-
redemptive-reconstruction-across-cold-war-borders-1st-edition-
demshuk/

Archaeological Conservation Using Polymers Practical

Applications for Organic Artifact Stabilization Texas A
M University Anthropology Series 1st Edition C. Wayne
Smith
https://ebookname.com/product/archaeological-conservation-using-
polymers-practical-applications-for-organic-artifact-
stabilization-texas-a-m-university-anthropology-series-1st-
edition-c-wayne-smith/

EU Shipping Law 3rd Edition Vincent Power

https://ebookname.com/product/eu-shipping-law-3rd-edition-
vincent-power/

Relational database index design and the optimizers DB2

Oracle SQL server et al 1st Edition Tapio Lahdenmaki

https://ebookname.com/product/relational-database-index-design-
and-the-optimizers-db2-oracle-sql-server-et-al-1st-edition-tapio-
lahdenmaki/
Electrical Conduction in Graphene and Nanotubes 1st
Edition Fujita

https://ebookname.com/product/electrical-conduction-in-graphene-
and-nanotubes-1st-edition-fujita/
 Basic Structured Grid Generation
with an introduction to unstructured grid generation
Basic Structured Grid
Generation
with an introduction to unstructured
grid generation

M. Farrashkhalvat and J.P. Miles

OXFORD AMSTERDAM BOSTON LONDON NEW YORK PARIS

SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY TOKYO
Butterworth-Heinemann
An imprint of Elsevier Science
Linacre House, Jordan Hill, Oxford OX2 8DP
200 Wheeler Rd, Burlington MA 01803

First published 2003

Copyright

c 2003, M. Farrashkhalvat and J.P. Miles. All rights reserved

The right of M. Farrashkhalvat and J.P. Miles to be identified as the authors of

this work has been asserted in accordance with the Copyright, Designs
and Patents Act 1988

No part of this publication may be

reproduced in any material form (including
photocopying or storing in any medium by electronic
means and whether or not transiently or incidentally
to some other use of this publication) without the
written permission of the copyright holder except
in accordance with the provisions of the Copyright,
Designs and Patents Act 1988 or under the terms of a
licence issued by the Copyright Licensing Agency Ltd,
90 Tottenham Court Road, London, England W1T 4LP.
Applications for the copyright holder’s written permission
to reproduce any part of this publication should be
addressed to the publisher

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication Data

A catalogue record for this book is available from the Library of Congress

ISBN 0 7506 5058 3

For information on all Butterworth-Heinemann publications visit our website at www.bh.com

Typeset by Laserwords Private Limited, Chennai, India

Printed and bound in Great Britain
 Contents

Preface ix

1. Mathematical preliminaries – vector and tensor analysis 1

1.1 Introduction 1
1.2 Curvilinear co-ordinate systems and base vectors in E 3 1
1.3 Metric tensors 4
1.4 Line, area, and volume elements 8
1.5 Generalized vectors and tensors 8
1.6 Christoffel symbols and covariant differentiation 14
1.7 Div, grad, and curl 19
1.8 Summary of formulas in two dimensions 23
1.9 The Riemann-Christoffel tensor 26
1.10 Orthogonal curvilinear co-ordinates 27
1.11 Tangential and normal derivatives – an introduction 28

2. Classical differential geometry of space-curves 30

2.1 Vector approach 30
2.2 The Serret-Frenet equations 32
2.3 Generalized co-ordinate approach 35
2.4 Metric tensor of a space-curve 38

3. Differential geometry of surfaces in E3 42

3.1 Equations of surfaces 42
3.2 Intrinsic geometry of surfaces 46
3.3 Surface covariant differentiation 51
3.4 Geodesic curves 54
3.5 Surface Frenet equations and geodesic curvature 57
3.6 The second fundamental form 60
3.7 Principal curvatures and lines of curvature 63
3.8 Weingarten, Gauss, and Gauss-Codazzi equations 67
3.9 Div, grad, and the Beltrami operator on surfaces 70
vi Contents

4. Structured grid generation – algebraic methods 76

4.1 Co-ordinate transformations 76
4.2 Unidirectional interpolation 80
4.2.1 Polynomial interpolation 80
4.2.2 Hermite interpolation polynomials 85
4.2.3 Cubic splines 87
4.3 Multidirectional interpolation and TFI 92
4.3.1 Projectors and bilinear mapping in two dimensions 92
4.3.2 Numerical implementation of TFI 94
4.3.3 Three-dimensional TFI 96
4.4 Stretching transformations 98
4.5 Two-boundary and multisurface methods 103
4.5.1 Two-boundary technique 103
4.5.2 Multisurface transformation 104
4.5.3 Numerical implementation 106
4.6 Website programs 108
4.6.1 Subdirectory: Book/univariate.gds 109
4.6.2 Subdirectory: Book/Algebra 109
4.6.3 Subdirectory: Book/bilinear.gds 112
4.6.4 Subdirectory: Book/tfi.gds 114
4.6.5 Subdirectory: Book/analytic.gds 115

5. Differential models for grid generation 116

5.1 The direct and inverse problems 116
5.2 Control functions 119
5.3 Univariate stretching functions 120
5.3.1 Orthogonality considerations 121
5.4 Conformal and quasi-conformal mapping 122
5.5 Numerical techniques 125
5.5.1 The Thomas Algorithm 125
5.5.2 Jacobi, Gauss-Seidel, SOR methods 127
5.5.3 The conjugate gradient method 129
5.6 Numerical solutions of Winslow equations 131
5.6.1 Thomas Algorithm 132
5.6.2 Orthogonality 134
5.7 One-dimensional grids 136
5.7.1 Grid control 136
5.7.2 Numerical aspects 139
5.8 Three-dimensional grid generation 140
5.9 Surface-grid generation model 141
5.10 Hyperbolic grid generation 142
5.11 Solving the hosted equations 143
5.11.1 An example 143
5.11.2 More general steady-state equation 145
5.12 Multiblock grid generation 146
5.13 Website programs 148
5.13.1 Subdirectory: Book/Winslow.gds 148
 Contents vii

5.13.2 Subdirectory: Book/one.d.gds 150

5.13.3 Subdirectory: Book/hyper.gds 150
5.13.4 Subdirectory: Book/p.d.Equations 151

6. Variational methods and adaptive grid generation 152

6.1 Introduction 152
6.2 Euler-Lagrange equations 153
6.3 One-dimensional grid generation 157
6.3.1 Variational approach 157
6.3.2 Dynamic adaptation 159
6.3.3 Space-curves 161
6.4 Two-dimensional grids 164
6.4.1 The L-functional and the Winslow model 165
6.4.2 The weighted L-functional 166
6.4.3 The weighted area-functional 167
6.4.4 Orthogonality-functional 167
6.4.5 Combination of functionals 168
6.4.6 Other orthogonality functionals 169
6.4.7 The Liao functionals 170
6.4.8 Surface grids 171
6.5 Harmonic maps 172
6.5.1 Surface grids 175
6.6 Website programs 177
6.6.1 Subdirectory: Book/var.gds 177
6.6.2 Subdirectory: Book/one.d.gds 179

7. Moving grids and time-dependent co-ordinate systems 180

7.1 Time-dependent co-ordinate transformations 180
7.2 Time-dependent base vectors 181
7.3 Transformation of generic convective terms 184
7.4 Transformation of continuity and momentum equations 185
7.4.1 Continuity equation 185
7.4.2 Momentum equations 185
7.5 Application to a moving boundary problem 187

8. Unstructured grid generation 190

8.1 Introduction 190
8.2 Delaunay triangulation 191
8.2.1 Basic geometric properties 191
8.2.2 The Bowyer-Watson algorithm 193
8.2.3 Point insertion strategies 196
8.3 Advancing front technique (AFT) 203
8.3.1 Introduction 203
8.3.2 Grid control 204
8.3.3 Searching algorithm 205
8.3.4 AFT algorithm 206
viii Contents

8.3.5 Adaptation and parameter space 216

8.3.6 Grid quality improvement 216
8.4 Solving hosted equations using finite elements 217
8.5 Website programs 221
8.5.1 Subdirectory: book/Delaunay 221

Bibliography 227

Index 229
 Preface

Over the past two decades, efficient methods of grid generation, together with the
power of modern digital computers, have been the key to the development of numer-
ical finite-difference (as well as finite-volume and finite-element) solutions of linear
and non-linear partial differential equations in regions with boundaries of complex
shape. Although much of this development has been directed toward fluid mechanics
problems, the techniques are equally applicable to other fields of physics and engi-
neering where field solutions are important. Structured grid generation is, broadly
speaking, concerned with the construction of co-ordinate systems which provide co-
ordinate curves (in two dimensions) and co-ordinate surfaces (in three dimensions)
that remain coincident with the boundaries of the solution domain in a given problem.
Grid points then arise in the interior of the solution domain at the intersection of these
curves or surfaces, the grid cells, lying between pairs of intersecting adjacent curves
or surfaces, being generally four-sided figures in two dimensions and small volumes
with six curved faces in three dimensions.
It is very helpful to have a good grasp of the underlying mathematics, which is
principally to be found in the areas of differential geometry (of what is now a fairly
old-fashioned variety) and tensor analysis. We have tried to present a reasonably self-
contained account of what is required from these subjects in Chapters 1 to 3. It is
hoped that these chapters may also serve as a helpful source of background reference
equations.
The following two chapters contain an introduction to the basic techniques (mainly
in two dimensions) of structured grid generation, involving algebraic methods and dif-
ferential models. Again, in an attempt to be reasonably inclusive, we have given a
brief account of the most commonly-used numerical analysis techniques for interpo-
lation and for solving algebraic equations. The differential models considered cover
elliptic and hyperbolic partial differential equations, with particular reference to the
use of forcing functions for the control of grid-density in the solution domain. For
solution domains with complex geometries, various techniques are used in practice,
including the multi-block method, in which a complex solution domain is split up
into simpler sub-domains. Grids may then be generated in each sub-domain (using the
sort of methods we have presented), and a matching routine, which reassembles the
sub-domains and matches the individual grids at the boundaries of the sub-domains, is
used. We show a simple matching routine at the end of Chapter 5.
A number of variational approaches (preceded by a short introduction to variational
methods in general) are presented in Chapter 6, showing how grid properties such
x Preface

as smoothness, orthogonality, and grid density can be controlled by the minimization

of an appropriate functional (dependent on the components of a fundamental metric
tensor). Surface grid generation has been considered here in the general context of
harmonic maps. In Chapter 7 time-dependent problems with moving boundaries are
considered. Finally, Chapter 8 provides an introduction to the currently very active area
of unstructured grid generation, presenting the fundamentals of Delaunay triangulation
and advancing front techniques.
Our aim throughout is to provide a straightforward and compact introduction to grid
generation, covering the essential mathematical background (in which, in our view,
tensor calculus forms an important part), while steering a middle course regarding the
level of mathematical difficulty. Mathematical exercises are suggested from time to
time to assist the reader. In addition, the companion website (www.bh.com/companions/
0750650583) provides a series of easy-to-follow, clearly annotated numerical codes,
closely associated with Chapters 4, 5, 6, and 8. The aim has been to show the applica-
tion of the theory to the generation of numerical grids in fairly simple two-dimensional
domains, varying from rectangles, circles and ellipses to more complex geometries,
such as C-grids over an airfoil, and thus to offer the reader a basis for further progress
in this field. Programs involve some of the most frequently used and familiar stable
numerical techniques, such as the Thomas Algorithm for the solution of tridiagonal
matrix equations, the Gauss-Seidel method, the Conjugate Gradient method, Succes-
sive Over Relaxation (SOR), Successive Line Over Relaxation, and the Alternating
Direction Implicit (ADI) method, as well as Transfinite Interpolation and the marching
algorithm (a grid generator for hyperbolic partial differential equations). The program-
ming language is the standard FORTRAN 77/90.
Our objective in this book is to give an introduction to the most important
aspects of grid generation. Our coverage of the literature is rather select-
ive, and by no means complete. For further information and a much wider
range of references, texts such as Carey (1997), Knupp and Steinberg (1993),
Thompson, Warsi, and Mastin (1985), and Liseikin (1999) may be consulted. Unstruc-
tured grid generation is treated in George (1991). A very comprehensive survey of mod-
ern developments, together with a great deal of background information, is provided
by Thompson, Soni, and Weatherill (1999).
The authors would like to express their gratitude to Mr. Thomas Sippel-Dau, LINUX
Service Manager at Imperial College of Science, Technology and Medicine for help
with computer administration.
M. Farrashkhalvat

J.P. Miles
 1

Mathematical
preliminaries – vector and tensor
analysis

1.1 Introduction
In this chapter we review the fundamental results of vector and tensor calculus which
form the basis of the mathematics of structured grid generation. We do not feel it
necessary to give derivations of these results from the perspective of modern dif-
ferential geometry; the derivations provided here are intended to be appropriate to
the background of most engineers working in the area of grid generation. Helpful
introductions to tensor calculus may be found in Kay (1988), Kreyzig (1968), and
Spain (1953), as well as many books on continuum mechanics, such as Aris (1962).
Nevertheless, we have tried to make this chapter reasonably self-contained. Some of
the essential results were presented by the authors in Farrashkhalvat and Miles (1990);
this book started at an elementary level, and had the restricted aim, compared with
many of the more wide-ranging books on tensor calculus, of showing how to use
tensor methods to transform partial differential equations of physics and engineer-
ing from one co-ordinate system to another (an aim which remains relevant in the
present context). There are some minor differences in notation between the present
book and Farrashkhalvat and Miles (1990).

1.2 Curvilinear co-ordinate systems and base

vectors in E3
We consider a general set of curvilinear co-ordinates x i , i = 1, 2, 3, by which points
in a three-dimensional Euclidean space E 3 may be specified. The set {x 1 , x 2 , x 3 }
could stand for cylindrical polar co-ordinates {r, θ, z}, spherical polars {r, θ, ϕ}, etc.
A special case would be a set of rectangular cartesian co-ordinates, which we shall
generally denote by {y1 , y2 , y3 } (where our convention of writing the integer indices
as subscripts instead of superscripts will distinguish cartesian from other systems),
or sometimes by {x, y, z} if this would aid clarity. Instead of {x 1 , x 2 , x 3 }, it may
occasionally be clearer to use notation such as {ξ, η, ς} without indices.
2 Basic Structured Grid Generation

The position vector r of a point P in space with respect to some origin O may be
expressed as
r = y1 i1 + y2 i2 + y3 i3 , (1.1)

where {i1 , i2 , i3 }, alternatively written as {i, j, k}, are unit vectors in the direction of the
rectangular cartesian axes. We assume that there is an invertible relationship between
this background set of cartesian co-ordinates and the set of curvilinear co-ordinates, i.e.

yi = yi (x 1 , x 2 , x 3 ), i = 1, 2, 3, (1.2)

with the inverse relationship

x i = x i (y1 , y2 , y3 ), i = 1, 2, 3. (1.3)

We also assume that these relationships are differentiable. Differentiating eqn (1.1)
with respect to x i gives the set of covariant base vectors
∂r
gi = , i = 1, 2, 3, (1.4)
∂x i
with background cartesian components
∂yj
(gi )j = , j = 1, 2, 3. (1.5)
∂x i
At any point P each of these vectors is tangential to a co-ordinate curve passing
through P, i.e. a curve on which one of the x i s varies while the other two remain
constant (Fig. 1.1). In general the gi s are neither unit vectors nor orthogonal to each
other. But so that they may constitute a set of basis vectors for vectors in E 3 we demand
that they are not co-planar, which is equivalent to requiring that the scalar triple product
{g1 · (g2 × g3 )}
= 0. Furthermore, this condition is equivalent to the requirement that
the Jacobian of the transformation (1.2), i.e. the determinant of the matrix of partial
derivatives (∂yi /∂x j ), is non-zero; this condition guarantees the existence of the inverse
relationship (1.3).

x 3 varies
g2
y3 g3 x 2 varies

P
g1
x1 varies
i3
i2 y2
O
i1

y1

Fig. 1.1 Covariant base vectors at a point P in three dimensions.

 Mathematical preliminaries – vector and tensor analysis 3

Given the set {g1 , g2 , g3 } we can form the set of contravariant base vectors at P,
{g1 , g2 , g3 }, defined by the set of scalar product identities
gi · gj = δji (1.6)

where δji is the Kronecker symbol given by

1 when i = j
δj =
i
(1.7)
0 when i
= j

Exercise 1. Deduce from the definitions (1.6) that the gi s may be expressed in terms
of vector products as
g2 × g3 g3 × g1 g1 × g2
g1 = , g2 = , g3 = (1.8)
V V V
where V = {g1 · (g2 × g3 )}. (Note that V represents the volume of a parallelepiped
(Fig. 1.2) with sides g1 , g2 , g3 .)

The fact that g1 is perpendicular to g2 and g3 , which are tangential to the co-ordinate
curves on which x 2 and x 3 , respectively, vary, implies that g1 must be perpendicular
to the plane which contains these tangential directions; this is just the tangent plane to
the co-ordinate surface at P on which x 1 is constant. Thus gi must be normal to the
co-ordinate surface x i = constant.
Comparison between eqn (1.6), with the scalar product expressed in terms of carte-
sian components, and the chain rule
∂x i ∂y1 ∂x i ∂y2 ∂x i ∂y3 ∂x i ∂yk ∂x i
+ + = = = δji (1.9)
∂y1 ∂x j ∂y2 ∂x j ∂y3 ∂x j ∂yk ∂x j ∂x j
for partial derivatives shows that the background cartesian components of gi are
given by
∂x i
(gi )j = , j = 1, 2, 3. (1.10)
∂yj
In eqn (1.9) we have made use of the summation convention, by which repeated
indices in an expression are automatically assumed to be summed over their range

g2
g3
V

g1

Fig. 1.2 Parallelepiped of base vectors at point P.

4 Basic Structured Grid Generation

of values. (In expressions involving general curvilinear co-ordinates the summation

convention applies only when one of the repeated indices appears as a subscript and
the other as a superscript.) The comparison shows that
∂x i ∂x i ∂x i
i1 + i2 + i3 = ∇x i = gi , (1.11)
∂y1 ∂y2 ∂y3
where the gradient operator ∇, or grad, is defined in cartesians by
∂ ∂ ∂ ∂
∇ = i1 + i2 + i3 = ik . (1.12)
∂y1 ∂y2 ∂y3 ∂yk
For a general scalar field ϕ we have
 
∂ϕ ∂ϕ ∂x j ∂x j ∂ϕ ∂ϕ
∇ϕ = ik = ik j = ik j
= gj j , (1.13)
∂yk ∂x ∂yk ∂yk ∂x ∂x
making use of a chain rule again and eqn (1.11); this gives the representation of the
gradient operator in general curvilinear co-ordinates.

1.3 Metric tensors

Given a set of curvilinear co-ordinates {x i } with covariant base vectors gi and con-
travariant base vectors gi , we can define the covariant and contravariant metric tensors
respectively as the scalar products
gij = gi · gj (1.14)
g ij = gi · gj , (1.15)
where i and j can take any values from 1 to 3. From eqns (1.5), (1.10), for the back-
ground cartesian components of gi and gi , it follows that
∂yk ∂yk
gij = (1.16)
∂x i ∂x j
and
∂x i ∂x j
g ij = . (1.17)
∂yk ∂yk
If we write (x, y, z) for cartesians and (ξ, η, ς) for curvilinear co-ordinates, we have
the formulas
g11 = xξ2 + yξ2 + zξ2
g22 = xη2 + yη2 + zη2
g33 = xς2 + yς2 + zς2 (1.18)
g12 = g21 = xξ xη + yξ yη + zξ zη
 Mathematical preliminaries – vector and tensor analysis 5

g23 = g32 = xη xς + yη yς + zη zς
g31 = g13 = xς xξ + yς yξ + zς zξ
∂x
where a typical partial derivative ∂ξ has been written as xξ , and the superscript 2 now
represents squaring.

Exercise 2. For the case of spherical polar co-ordinates, with ξ = r, η = θ , ς = ϕ, and

x = r sin θ cos ϕ, y = r sin θ sin ϕ, z = r cos θ
show that    
g11 g12 g13 1 0 0
 g21 g22 g23  =  0 r 2 0 , (1.19)
g31 g32 g33 2 2
0 0 r sin θ
where (r, θ, ϕ) take the place of (ξ, η, ς).

Formulas for g ij are, similarly,

g 11 = ξx2 + ξy2 + ξz2
g 22 = ηx2 + ηy2 + ηz2
g 33 = ςx2 + ςy2 + ςz2 (1.20)
g 12 = g 21 = ξx ηx + ξy ηy + ξz ηz
g 23 = g 32 = ηx ςx + ηy ςy + ηz ςz
g 31 = g 13 = ςx ξx + ςy ξy + ςz ξz .
The metric tensor gij provides a measure of the distance ds between neighbouring
points. If the difference in position vectors between the two points is dr and the
infinitesimal differences in curvilinear co-ordinates are dx 1 , dx 2 , dx 3 , then
 3  
3
∂r ∂r ∂r ∂r
ds 2 = dr · dr = dx i ·  dx j  = i · j dx i dx j = gij dx i dx j ,
∂x i ∂x j ∂x ∂x
i=1 j =1
(1.21)
making use of the summation convention. As previously remarked, the summation
convention may be employed in generalized (curvilinear) co-ordinates only when each
of the repeated indices appears once as a subscript and once as a superscript.
We can form the 3 × 3 matrix L whose row i contains the background cartesian
components of gi and the matrix M whose row i contains the background cartesian
components of gi . We may write, in shorthand form,
 
g1
L =  g2  , M T = g1 g2 g3 (1.22)
g3
6 Basic Structured Grid Generation

and Lij =
∂yj ∂x i
i , Mij = ∂yj ; it may be seen directly from eqn (1.6) that
∂x
LM T = I, (1.23)

where I is the 3 × 3 identity matrix. Thus L and M T are mutual inverses. Moreover

det L = {g1 · (g2 × g3 )} = V (1.24)

as previously defined in eqn (1.8). Since M T = L−1 , it follows that

det M = {g1 · (g2 × g3 )} = V −1 . (1.25)

It is easy to see that the symmetric matrix arrays (gij ) and (g ij ) for the associated
metric tensors are now given by

(gij ) = LLT , (g ij ) = MM T . (1.26)

Since M T = L−1 and M = (LT )−1 , it follows that

(g ij ) = (gij )−1 . (1.27)

In component form this is equivalent to

j
gik g j k = δi . (1.28)

From the properties of determinants it also follows that

g = det(gij ) = (det L)2 = V 2 , (1.29)

det(g ij ) = g −1 , (1.30)

and
√
V = {g1 · (g2 × g3 )} = g, (1.31)

where g must be a positive quantity.

Thus in place of eqn (1.8) we can write
1 1 1
g1 = √ g2 × g3 , g2 = √ g3 × g1 , g3 = √ g1 × g2 . (1.32)
g g g
From eqn (1.27) and standard 3 × 3 matrix inversion, we can also deduce the fol-
lowing formula:  
1 G1 G4 G5
g ij =  G4 G2 G6  , (1.33)
g G5 G6 G3

where the co-factors of (gij ) are given by

G1 = g22 g33 − (g23 )2 , G2 = g11 g33 − (g13 )2 , G3 = g11 g22 − (g12 )2

G4 = g13 g23 − g12 g33 , G5 = g12 g23 − g13 g22 , G6 = g12 g13 − g23 g11 . (1.34)
 Mathematical preliminaries – vector and tensor analysis 7

The cofactors of the matrix L in eqn (1.22) are the various background carte-
sian components of (gj × gk ), which may be expressed, with the notation used in
eqn (1.18), as
α1 = yη zς − yς zη , α2 = xς zη − xη zς , α3 = xη yς − xς yη
β1 = yς zξ − yξ zς , β2 = xξ zς − xς zξ , β3 = xς yξ − xξ yς (1.35)
γ1 = yξ zη − yη zξ , γ2 = xη zξ − xξ zη , γ3 = xξ yη − xη yξ
so that
g2 × g3 = α1 i + α2 j + α3 k, g3 × g1 = β1 i + β2 j + β3 k, g1 × g2 = γ1 i + γ2 j + γ3 k
(1.36)
and
1 1 1
g1 = √ (α 1 i+α2 j+α3 k), g2 = √ (β1 i+β2 j+β3 k), g3 = √ (γ1 i+γ2 j+γ3 k).
g g g
(1.37)
Since M = L−1 , we also have, in the same notation, the matrix elements of M:
√ √ √
ξx = α1 / g, ξy = α2 / g, ξz = α3 / g
√ √ √
ηx = β1 / g, ηy = β2 / g, ηz = β3 / g (1.38)
√ √ √
ςx = γ1 / g, ςy = γ2 / g, ςz = γ3 / g.

Exercise 3. Using eqn (1.29) and standard determinant expansions, derive the follow-
ing formulas for the determinant g:
g = g11 G1 + g12 G4 + g13 G5 = (α1 xξ + β1 xη + γ1 xς )2
= g22 G2 + g12 G4 + g23 G6 = (α2 yξ + β2 yη + γ2 yς )2 (1.39)
= g33 G3 + g13 G5 + g23 G6 = (α3 zξ + β3 zη + γ3 zς )2 .

From eqn (1.32) it follows that

1
g ip = gi · gp = (gj × gk ) · (gq × gr ),
g
where {i, j, k} and {p, q, r} are in cyclic order {1, 2, 3}. Using the standard Lagrange
vector identity
(A × B) · (C × D) = (A · C)(B · D) − (A · D)(B · C), (1.40)
we have
1
g ip = {(gj · gq )(gk · gr ) − (gj · gr )(gk · gq )}
g
1
= (gj q gkr − gj r gkq ). (1.41)
g
For example,
1
g 13 = (g21 g32 − g22 g31 ).
g
8 Basic Structured Grid Generation

1.4 Line, area, and volume elements

Lengths of general infinitesimal line-elements are given by eqn (1.21). An element of
the x 1 co-ordinate curve on which dx 2 = dx 3 = 0 is therefore given by (ds)2 =
g11 (dx 1 )2 . Thus arc-length along the x i -curve is
√
ds = gii dx i (1.42)
(with no summation over i).
A line-element along the x 1 -curve may be written ∂x ∂r 1 = g dx 1 , and simi-
1 dx 1
2 2
larly a line-element along the x -curve is g2 dx . The infinitesimal vector area of the
parallelogram of which these two line-elements form the sides is the vector product
(g1 dx 1 × g2 dx 2 ), which has magnitude
dA3 = |g1 × g2 | dx 1 dx 2 . (1.43)
Again by the Lagrange vector identity we have
|g1 × g2 |2 = (g1 × g2 ) · (g1 × g2 ) = (g1 · g1 )(g2 · g2 ) − (g1 · g2 )(g1 · g2 )
= g11 g22 − (g12 )2 .

Hence dA3 = g11 g22 − (g12 )2 dx 1 dx 2 , giving the general expression


dAi = gjj gkk − (gj k )2 dx j dx k = Gi dx j dx k , (1.44)

using eqn (1.34), where i, j, k must be taken in cyclic order 1, 2, 3, and again there is
no summation over j and k.
The parallelepiped generated by line-elements g1 dx 1 , g2 dx 2 , g3 dx 3 , along the co-
ordinate curves has infinitesimal volume
dV = g1 dx 1 · (g2 dx 2 × g3 dx 3 ) = {g1 · (g2 × g3 )}dx 1 dx 2 dx 3 .
By eqn (1.31) we have
√
dV = g dx 1 dx 2 dx 3 . (1.45)

1.5 Generalized vectors and tensors

A vector field u (a function of position r) may be expressed at a point P in terms of
the covariant base vectors g1 , g2 , g3 , or in terms of the contravariant base vectors g1 ,
g2 , g3 . Thus we have
u = u1 g1 + u2 g2 + u3 g3 = ui gi (1.46)
= u1 g + u2 g + u3 g = ui g ,
1 2 3 i
(1.47)
where ui and ui are called the contravariant and covariant components of u, respect-
ively. Taking the scalar product of both sides of eqn (1.46) with gj gives
j
u · gj = ui gi · gj = ui δi = uj .
 Mathematical preliminaries – vector and tensor analysis 9

Hence
ui = u · gi , (1.48)
and, similarly,
ui = u · gi . (1.49)
A similar procedure shows, incidentally, that
gi = gij gj , (1.50)
and
gi = g ij gj . (1.51)
We then easily deduce that
ui = g ij uj (1.52)
and
ui = gij uj . (1.53)

These equations may be interpreted as demonstrating that the action of g ij on uj

and that of gij on uj are effectively equivalent to ‘raising the index’ and ‘lowering the
index’, respectively.
It is straightforward to show that the scalar product of vectors u and v is given by
u · v = ui vi = ui v i = gij ui v j = g ij ui vj (1.54)
and hence that the magnitude of a vector u is given by
 
|u| = gij ui uj = g ij ui uj . (1.55)

It is important to note the special transformation properties of covariant and con-

travariant components under a change of curvilinear co-ordinate system. We consider
another system of co-ordinates x i , i = 1, 2, 3, related to the first system by the trans-
formation equations
x i = x i (x 1 , x 2 , x 3 ), i = 1, 2, 3. (1.56)
These equations are assumed to be invertible and differentiable. In particular, dif-
ferentials in the two systems are related by the chain rule
∂x i
dx i = dx j , (1.57)
∂x j
or, in matrix terms,    
dx 1 dx 1
 dx 2  = A  dx 2  , (1.58)
dx 3 dx 3
where we assume that the matrix A of the transformation, with i-j element equal to
∂x i /∂x j , has a determinant not equal to zero, so that eqn (1.58) may be inverted. We
define the Jacobian J of the transformation as
J = det A. (1.59)
10 Basic Structured Grid Generation

Exercise 4. Show that if we define the matrix B as that whose i-j element is equal to
∂x j /∂x i , then
AB T = I (1.60)

and
det B = J −1 . (1.61)

We obtain new covariant base vectors, which transform according to the rule
∂r ∂r ∂x j ∂x j
gi = = = gj , (1.62)
∂x i ∂x j ∂x i ∂x i
with the inverse relationship
∂x j
gi = g . (1.63)
∂x i j
In background cartesian components, eqn (1.62) may be written in matrix form as

L = BL, (1.64)

where L is the matrix with i-j component given by ∂yj /∂x i , and from eqn (1.23) and
eqn (1.60) we deduce that
M = AM, (1.65)

where M is the matrix with i-j component ∂x i /∂yj .

The new system of co-ordinates has associated metric tensors given, in comparison
with eqn (1.26), by
T T
(g ij ) = L L , (g ij ) = M M , (1.66)

so that the corresponding determinant g = det(g ij ) is given by

g = (det L)2 .
√ √
Hence det L = g, det L = g, and det L = det B det L = J −1 det L from
eqn (1.64). Thus we have 
g
J = . (1.67)
g
Equation (1.65) yields the relation between corresponding contravariant base vectors:

∂x i j
gi = g . (1.68)
∂x j
Expressing u as a linear combination of the base vectors in the new system gives

u = ui gi = ui gi . (1.69)

We now easily obtain, using eqn (1.62), the transformation rule for the covariant
components of a vector:
 j 
∂x ∂x j ∂x j
ui = u·gi = u· gj = u · gj = uj , (1.70)
∂x i ∂x i ∂x i
 Mathematical preliminaries – vector and tensor analysis 11

or, in matrix form,    

u1 u1
 u2  = B  u2  . (1.71)
u3 u3
∂ϕ
The set of components ∂x j (where ϕ is a scalar field) found in eqn (1.13) can be said
to constitute a covariant vector, since by the usual chain rule they transform according
to eqn (1.70), i.e.
∂ϕ ∂x j ∂ϕ
= .
∂x i ∂x i ∂x j
Exercise 5. Show that the transformation rule for contravariant components of a
vector is
∂x i
ui = j uj , (1.72)
∂x
or  1   1 
u u
 u2  = A  u2  . (1.73)
u3 u3

Note the important consequence that the scalar product (1.54) is an invariant quantity
(a true scalar), since it is unaffected by co-ordinate transformations. In fact
  k   i k
∂x i j ∂x ∂x ∂x
u vi =
i
u v =
i k
uj vk = δjk uj vk = uj vj .
∂x j ∂x ∂x j
∂x i

From eqns (1.64), (1.65), and (1.66), we obtain the transformation rules:
T
(g ij ) = L L = BLLT B T = B(gij )B T , (1.74)
T
(g ij ) = M M = AMM T AT = A(g ij )AT . (1.75)

In fact gij is a particular case of a covariant tensor of order two, which may be
defined here as a set of quantities which take the values Tij , say, when the curvilinear
co-ordinates x i are chosen and the values T ij when a different set x i are chosen, with a
transformation rule between the two sets of values being given in co-ordinate form by
∂x k ∂x l
T ij = Tkl (1.76)
∂x i ∂x j
with summation over k and l, or in matrix form

T = BT B T . (1.77)

Similarly, g ij is a particular case of a contravariant tensor of order two. This is

defined as an entity which has components T ij obeying the transformation rules

ij ∂x i ∂x j kl
T = T (1.78)
∂x k ∂x l
12 Basic Structured Grid Generation

or, equivalently,
T = ATAT . (1.79)
.j j
We can also define mixed second-order tensors Ti and T.i , for which the transfor-
mation rules are
.j ∂x k ∂x j .l
Ti = Tk (1.80)
∂x i ∂x l
T = BTAT , (1.81)
and
i ∂x i ∂x l k
T ..j = T (1.82)
∂x k ∂x j .l
T = ATBT . (1.83)

Exercise 6. Show from the transformation rules (1.80) and (1.82) that the quantities
T.kk and Tk.k are invariants.

Given two vectors u and v, second-order tensors can be generated by taking products
of covariant or contravariant vector components, giving the covariant tensor ui vj , the
contravariant tensors ui v j , and the mixed tensors ui vj and ui v j . In this case these
tensors are said to be associated, since they are all derived from an entity which
can be written in absolute, co-ordinate-free, terms, as u ⊗ v; this is called the dyadic
product of u and v. The dyadic product may also be regarded as a linear operator
which acts on vectors w according to the rule
(u ⊗ v)w = u(v · w), (1.84)
an equation which has various co-ordinate representations, such as

(ui vj )wj = ui (vj wj ).

It may also be expressed in the following various ways:

u ⊗ v = ui vj gi ⊗ gj = ui v j gi ⊗ gj = ui vj gi ⊗ gj = ui v j gi ⊗ gj (1.85)
with summation over i and j in each case.
.j
In general, covariant, contravariant, and mixed components Tij , T ij , Ti , T.ji , are
associated if there exists an entity T, a linear operator which can operate on vectors,
such that
..j
T = Tij gi ⊗ gj = T ij gi ⊗ gj = Ti gi ⊗ gj = T..j
i
gi ⊗ gj . (1.86)
Thus the action of T on a vector u could be represented typically by

Tu = (Tij gi ⊗ gj )u = Tij gi (gj · u) = Tij gi uj = Tij uj gi = v,

where v has covariant components

vi = Tij uj .
 Mathematical preliminaries – vector and tensor analysis 13

The Kronecker symbol δji has corresponding matrix elements given by the 3 × 3
identity matrix I . It may be interpreted as a second-order mixed tensor, where which-
ever of the covariant or contravariant components occurs first is immaterial, since if we
substitute T = I in either of the transformation rules (1.81) or (1.83) we obtain T = I
in view of eqn (1.60). Thus δji is a mixed tensor which has the same components on any
co-ordinate system. The corresponding linear operator is just the identity operator I,
which for any vector u satisfies
Iu = (δji gi ⊗ gj )u = δji gi uj = gi ui = u.
The following representations of I may then be deduced:
I = gij gi ⊗ gj = g ij gi ⊗ gj = gj ⊗ gj = gj ⊗ gj . (1.87)

Thus gij , g ij , and δji are associated tensors.

Covariant, contravariant, and mixed tensors of higher order than two may be defined
in terms of transformation rules following the pattern in eqns (1.76), (1.78), (1.80), and
(1.82), though it may not be convenient to express these rules in matrix terms. For
example, covariant and contravariant third-order tensors Uijk and U ijk respectively
must follow the transformation rules:
∂x l ∂x m ∂x n ijk ∂x i ∂x j ∂x k lmn
U ijk = i j k
Ulmn , U = U . (1.88)
∂x ∂x ∂x ∂x l ∂x m ∂x n
The alternating symbol eijk defined by


 1 if (i, j, k) is an even permutation of (1, 2, 3)
eijk = e = −1
ijk
if (i, j, k) is an odd permutation of (1, 2, 3) (1.89)

 0 otherwise
is not a (generalized) third-order tensor. Applying the left-hand transformation of
eqns (1.88) gives, using the properties of determinants and eqns (1.61) and (1.67),

∂x l ∂x m ∂x n −1 g
i j
e
k lmn
= (det B)eijk = J eijk = eijk . (1.90)
∂x ∂x ∂x g
Similarly we obtain

∂x i ∂x j ∂x k lmn g ijk
e = (det A)eijk = J eijk = e . (1.91)
∂x l ∂x m ∂x n g
It follows that third-order covariant and contravariant tensors respectively are
defined by √
εijk = geijk (1.92)
and
1
εijk = √ eijk . (1.93)
g
√
Applying the appropriate transformation law to εijk now gives g elmn = εlmn ,
as required, and similarly for εijk . These tensors, known as the alternating tensors,
14 Basic Structured Grid Generation

are required, for example, when forming correct vector expressions in curvilinear co-
ordinate systems.
In particular, the vector product of two vectors u and v is given by
u × v = εijk uj vk gi = εijk uj v k gi , (1.94)
with summation over i, j, k. The component forms of the scalar triple product of
vectors u, v, w are
u · (v × w) = εijk ui vj wk = εijk ui v j wk . (1.95)
The alternating symbols themselves may be called relative (rather than absolute) ten-
sors, which means that when the tensor transformation law is applied as in eqns (1.90)
and (1.91) a power of J (the weight of the relative tensor) appears on the right-hand
side. Thus according to (1.90) eijk is a relative tensor of weight −1, while according
to eqn (1.91) eijk (although it takes exactly the same values as eijk ) is a relative tensor
of weight 1.

1.6 Christoffel symbols and covariant differentiation

In curvilinear co-ordinates the base vectors will generally vary in magnitude and direc-
tion from one point to another, and this causes special problems for the differentiation
of vector and tensor fields. In general, differentiation of covariant base vectors eqn (1.4)
with respect to x j satisfies
∂gi ∂ 2r ∂ 2r ∂gj
j
= j i
= i j
= . (1.96)
∂x ∂x ∂x ∂x ∂x ∂x i
Expressing the resulting vector (for a particular choice of i and j ) as a linear com-
bination of base vectors gives
∂gi
= [ij, k]gk = k
ij gk , (1.97)
∂x j
with summation over k. The coefficients [ij, k], ijk in eqn (1.97) are called Christoffel
symbols of the first and second kinds, respectively. Taking appropriate scalar products
on eqn (1.97) gives
∂gi
[ij, k] = j · gk (1.98)
∂x
and
∂gi
ij = · gk .
k
(1.99)
∂x j
Both [ij, k] and k
ij are symmetric in i and j by eqn (1.96). We also have, by
eqn (1.51),
∂gi
k
ij = · (g kl gl ) = g kl [ij, l] (1.100)
∂x j
with summation over l. Similarly,
[ij, k] = gkl l
ij . (1.101)
 Another Random Document on
Scribd Without Any Related Topics
 τὰ τῶν βαρβάρων ἱκανὸν εἶναι πράττειν.

169 Plato, Republic, vi. p. 494 D-E. ἐὰν δ’ οὖν, διὰ τὸ

εὖ πεφυκέναι καὶ τὸ ξυγγενὲς τῶν λόγων, εἶς
αἰσθάνηταί τέ πῃ καὶ κάμπτηται καὶ ἕλκηται πρὸς
φιλοσοφίαν, τί οἰόμεθα δράσειν ἐκείνους τοὺς
ἡγουμένους ἀπολλύναι αὐτοῦ τὴν χρείαν τε καὶ
ἑταιρείαν; οὐ πᾶν μὲν ἔργον, πᾶν δ’ ἔπος, λέγοντάς τε
καὶ πράττοντας καὶ περὶ αὐτόν, ὅπως ἂν μὴ πεισθῇ, καὶ
περὶ τὸν πείθοντα, ὅπως ἂν μὴ οἷός τ’ ᾖ, καὶ ἰδίᾳ
ἐπιβουλεύοντας καὶ δημοσίᾳ εἰς ἀγῶνας καθίσταντας;

170 Plato, Republic, vi. p. 495 A-B.

It is thus that the path of philosophy

The really great minds are
is deserted by those who ought to have
thus driven away from the
trodden it, and who pervert their
path of philosophy — which
exalted powers to unworthy objects.
is left to empty pretenders.
That path — being left vacant, yet still
full of imposing titles and pretensions, and carrying a show of
superior dignity as compared with the vulgar professions — becomes
invaded by interlopers of inferior worth and ability, who quit their
own small craft, and set up as philosophers.171 Such men, poorly
endowed by nature, and debased by habits of trade, exhibit
themselves, in their self-assumed exaltation as philosophers, like a
slave recently manumitted, who has put on new clothes and married
his master’s daughter.172 Having intruded themselves into a career
for which they are unfit, they cannot produce any grand or genuine
philosophical thoughts, or any thing better than mere neat sophisms,
pleasing to the ear.173 Through them arises the discredit which is
now attached to philosophers.
 171 Plato, Republic, vi. p. 495 C-D. καθορῶντες γὰρ
ἄλλοι ἀνθρωπίσκοι κενὴν τὴν χώραν ταύτην
γιγνομένην, καλῶν δὲ ὀνομάτων καὶ προσχημάτων
μεστήν, ὥσπερ οἱ ἐκ τῶν εἰργμῶν εἰς τὰ ἱερὰ
ἀποδιδράσκοντες, ἄσμενοι καὶ οὗτοι ἐκ τῶν τεχνῶν
ἐκπηδῶσιν εἰς τὴν φιλοσοφίαν.

172 Plato, Republic, vi. p. 495 E.

173 Plato, Republic, vi. p. 496 A.

Amidst such general degradation of

Rare cases in which a
philosophy, some few and rare cases
highly qualified philosopher
are left, in which the pre-eminent
remains — Being at
natures qualified for philosophy remain
variance with public
by some favourable accident
opinion, he can achieve
uncorrupted. One of these is Theagês,
nothing, and is lucky if he
who would have been long ago drawn
can obtain safety by
away from philosophy to active politics,
silence.
had he not been disqualified by bad
health. The restraining Dæmon, peculiar to myself (says Sokrates),
is another case.174 Such an exceptional man, having once tasted the
sweetness and happiness of philosophy, embraces it as an exclusive
profession. He sees that the mass of society are wrongheaded —
that scarce any one takes wholesome views on social matters — that
he can find no partisans to aid him in upholding justice175 — that
while he will not take part in injustice, he is too weak to contend
single-handed against the violence of all, and would only become a
victim to it without doing any good either to the city or to his friends
— like a man who has fallen among wild beasts. On these grounds
he stands aloof in his own separate pursuit, like one sheltering
himself under a wall against a hurricane of wind and dust.
Witnessing the injustice committed by all around, he is content if he
can keep himself clear and pure from it during his life here, so as to
die with satisfaction and good hopes.

174 Plato, Republic, vi. p. 496 D.

175 Plato, Republic, vi. p. 496 C-D. καὶ τούτων δὴ τῶν

ὀλίγων οἱ γευόμενοι καὶ γευσάμενοι ὡς ἡδὺ καὶ
μακάριον τὸ κτῆμα, καὶ τῶν πολλῶν αὖ ἱκανῶς ἰδόντες
τὴν μανίαν, καὶ ὅτι οὐδεὶς οὐδὲν ὑγιὲς, ὡς ἔπος εἰπεῖν,
περὶ τὰ τῶν πόλεων πράττει, οὐδ’ ἔστι ξύμμαχος μεθ’
ὅτου τις ἰὼν ἐπὶ τὴν τῶν δικαίων βοήθειαν σώζοιτ’ ἄν,
ἀλλ’ ὥσπερ εἰς θηρία ἄνθρωπος ἐμπεσών, οὔτε
ξυναδικεῖν ἐθέλων οὔτε ἱκανὸς ὢν εἷς πᾶσιν ἀγρίοις
ἀντέχειν, πρίν τι τὴν πόλιν ἢ φίλους ὀνησαι
προαπολόμενος ἀνωφελὴς αὑτῷ τε καὶ τοῖς ἄλλοις ἂν
γένοιτο — ταῦτα πάντα λογισμῷ λαβῶν, ἡσυχίαν ἔχων
καὶ τὰ αὐτοῦ πράττων … ὁρῶν τοὺς ἄλλους
καταπιμπλαμένους ἀνομίας, ἀγαπᾷ εἴ πη αὐτὸς καθαρὸς
ἀδικίας, &c.

He will perform no small achievement (remarks Adeimantus) if he

keeps clear to the end.176

176 Plato, Republic, vi. p. 497 A.

True (replies Sokrates) — yet

The philosopher must have
nevertheless he can perform no great
a community suitable to
achievement, unless he meets with a
him, and worthy of him.
community suited to him. Amidst such
a community he will himself rise to greatness, and will preserve the
public happiness as well as his own. But there exists no such
community anywhere, at the present moment. Not one of those now
existing is worthy of a philosophical disposition:177 which accordingly
becomes perverted, and degenerates into a different type adapted to
its actual abode, like exotic seed transported to a foreign soil. But if
this philosophical disposition were planted in a worthy community, so
as to be able to assert its own superior excellence, it would then
prove itself truly divine, leaving other dispositions and pursuits
behind as merely human.

177 Plato, Republic, vi. p. 497 B-C.

You mean by a worthy community

It must be such a
(observes Adeimantus), such an one as
community as Sokrates has
that of which you have been drawing
been describing — But
the outline? — I do (replies Sokrates):
means must be taken to
with this addition, already hinted but
keep up a perpetual
not explained, that there must always
succession of philosophers
be maintained in it a perpetual
as Rulers.
supervising authority representing the
scheme and purpose of the primitive lawgiver. This authority must
consist of philosophers: and the question now arises — difficult but
indispensable — how such philosophers are to be trained up and
made efficient for the good of the city.

The plan now pursued for imparting

Proper manner of teaching
philosophy is bad. Some do not learn it
philosophy — Not to begin
at all: and even to those who learn it
at a very early age.
best, the most difficult part (that which
relates to debate and discourse) is taught when they are youths just
emerging from boyhood, in the intervals of practical business and
money-getting.178 After that period, in their mature age, they
abandon it altogether; they will scarcely so much as go to hear an
occasional lecture on the subject, without any effort of their own:
accordingly it has all died out within them, when they become
mature in years. This manner of teaching philosophy ought to be
reversed. In childhood and youth, instruction of an easy character
and suitable to that age ought to be imparted; while the greatest
care is taken to improve and strengthen the body during its period of
growth, as a minister and instrument to philosophy. As age
proceeds, and the mind advances to perfection, the mental exercises
ought to become more difficult and absorbing. Lastly, when the age
of bodily effort passes away, philosophy ought to become the main
and principal pursuit.179

178 Plato, Republic, vi. p. 498 A. Νῦν μὲν οἱ καὶ

ἁπτόμενοι μειράκια ὄντα ἄρτι ἐκ παιδῶν τὸ μεταξὺ
οἰκονομίας καὶ χρηματισμοῦ πλησιάσαντες αὐτοῦ τῷ
χαλεπωτάτῳ ἀπαλλάττονται, οἱ φιλοσοφώτατοι
ποιούμενοι· λέγω δὲ χαλεπώτατον τὸ περὶ τοὺς
λόγους· ἐν δὲ τῷ ἔπειτα, ἐὰν καὶ ἄλλων τοῦτο
πραττόντων παρακαλούμενοι ἐθέλωσιν ἀκροαταὶ
γίγνεσθαι, μεγάλα ἡγοῦνται, πάρεργον οἰόμενοι αὐτὸ
δεῖν πράττειν.

179 Plato, Republic, vi. p. 498 C.

Most people will hear all this

If the multitude could once
(continues Sokrates) with mingled
see a real, perfect,
incredulity and repugnance. We cannot
philosopher, they could not
wonder that they do so: for they have
fail to love him: but this
had no experience of one or a few
never happens.
virtuously trained men ruling in a city
suitably prepared.180 Such combination of philosophical rulers within
a community adapted to them, we must assume to be realised.181
Though difficult, it is noway impracticable: and even the multitude
will become reconciled to it, if you explain to them mildly what sort
of persons we mean by philosophers. We do not mean such persons
as the multitude now call by that name; interlopers in the pursuit,
violent in dispute and quarrel with each other, and perpetually
talking personal scandal.182 The multitude cannot hate a
philosophical temper such as we depict, when they once come to
know it — a man who, indifferent to all party disputes, dwells in
contemplation of the Universal Forms, and tries to mould himself
and others into harmony with them.183 Such a philosopher will not
pretend to make regulations, either for a city or for an individual,
until he has purified it thoroughly. He will then make regulations
framed upon the type of the Eternal Forms — Justice, Temperance,
Beauty — adapting them as well as he can to human exigencies.184
The multitude, when they know what is really meant, will become
perfectly reconciled to it. One single prince, if he rises so as to
become a philosopher, and has a consenting community, will suffice
to introduce the system which we have been describing. So
fortunate an accident can undoubtedly occur but seldom; yet it is
not impossible, and one day or other it will really occur.185

180 Plato, Republic, vi. p. 498 E.

181 Plato, Republic, vi. p. 499 B-C.

182 Plato, Republic, vi. pp. 499-500.

183 Plato, Republic, vi. p. 500 C-D.

184 Plato, Republic, vi. p. 501 A.

185 Plato, Republic, vi. p. 502.

 I must now (continues Sokrates)
Course of training in the
explain more in detail the studies and
Platonic city, for imparting
training through which these
philosophy to the Rulers.
preservers Rulers of our city, the
They must be taught to
complete philosophers, must be
ascend to the Idea of Good.
created. The most perfect among the
But what is Good?
Guardians, after having been tested by
years of exercises and temptations of various kinds, will occupy that
distinguished place. Very few will be found uniting those distinct and
almost incompatible excellences which qualify them for the post.
They must give proof of self-command against pleasures as well as
pains, and of competence to deal with the highest studies.186 But
what are the highest studies? What is the supreme object of
knowledge? It is the Idea of Good — the Form of Good: to the
acquisition of which our philosophers must be trained to ascend,
however laborious and difficult the process may be.187 Neither
justice nor any thing else can be useful or profitable, unless we
superadd to them a knowledge of the Idea of Good: without this, it
would profit us nothing to possess all other knowledge.188

186 Plato, Republic, vi. p. 503.

187 Plato, Republic, vi. p. 504.

188 Plato, Republic, vi. p. 505 A. ὅτι γε ἡ τοῦ ἀγαθοῦ

ἰδέα μέγιστον μάθημα πολλάκις ἀκήκοας, ᾖ δίκαια καὶ
τἄλλα προσχρησάμενα χρήσιμα καὶ ὠφέλιμα γίγνεται,
&c.

Now as to the question, What Good

Ancient disputes upon this
is? there are great and long-standing
point, though every one
disputes. Every mind pursues Good,
yearns after Good. Some
say Intelligence; some say and does every thing for the sake of it
Pleasure. Neither is — yet without either knowledge or firm
satisfactory. assurance what Good is, and
consequently with perpetual failure in
deriving benefit from other acquisitions.189 Most people say that
Pleasure is the Good: an ingenious few identify Intelligence with the
Good. But neither of these explanations is satisfactory. For when a
man says that Intelligence is the Good, our next question to him
must be, What sort of Intelligence do you mean? — Intelligence of
what? To this he must reply, Intelligence of the Good: which is
absurd, since it presumes us to know already what the Good is —
the very point which he is pretending to elucidate. Again, he who
contends that Pleasure is the Good, is forced in discussion to admit
that there are such things as bad pleasures: in other words, that
pleasure is sometimes good, sometimes bad.190 From these doubts
and disputes about the real nature of good, we shall require our
philosophical Guardians to have emancipated themselves, and to
have attained a clear vision. They will be unfit for their post it they
do not well know what the Good is, and in what manner just or
honourable things come to be good.191 Our city will have received its
final consummation, when it is placed under the superintendence of
one who knows what the Good is.

189 Plato, Republic, vi. p. 505 E. Ὃ δὴ διώκει μὲν

ἅπασα ψυχὴ καὶ τούτου ἕνεκα πάντα πράττει,
ἀπομαντευομένη τὶ εἶναι, ἀποροῦσα δὲ καὶ οὐκ ἔχουσα
λαβεῖν ἱκανῶς τί ποτ’ ἐστίν, οὐδὲ πίστει χρήσασθαι
μονίμῳ, οἵᾳ καὶ περὶ τἄλλα, διὰ τοῦτο δὲ ἀποτυγχάνει
καὶ τῶν ἄλλων εἴ τι ὄφελος ἦν, &c.

190 Plato, Republic, vi. p. 505 C.

 191 Plato, Republic, vi. p. 506 A. δίκαιά τε καὶ καλὰ
ἀγνοούμενα ὅπῃ ποτὲ ἀγαθά ἐστιν, οὐ πολλοῦ τινὸς
ἄξιον φύλακα κεκτῆσθαι ἂν ἑαυτῶν τὸν τοῦτο
ἀγνοοῦντα.

But tell me, Sokrates (asks

Adeimantus asks what
Adeimantus), what do you conceive the
Sokrates says. Sokrates
Good to be — Intelligence or Pleasure,
says that he can not
or any other thing different from these?
answer: but he compares it
I do not profess to know (replies
by a metaphor to the Sun.
Sokrates), and cannot tell you. We
must decline the problem, What Good itself is? as more arduous
than our present impetus will enable us to reach.192 Nevertheless I
will partially supply the deficiency by describing to you the offspring
of Good, very like its parent. You will recollect that we have
distinguished the Many from the One: the many just particulars,
beautiful particulars, from the One Universal Idea or Form, Just per
se, Beautiful per se. The many particulars are seen but not
conceived: the one Idea is conceived, but not seen.193 We see the
many particulars through the auxiliary agency of light, which
emanates from the Sun, the God of the visible world. Our organ and
sense of vision are not the Sun itself, but they are akin to the Sun in
a greater degree than any of our other senses. They imbibe their
peculiar faculty from the influence of the Sun.194 The Sun furnishes
to objects the power of being seen, and to our eyes the power of
seeing: we can see no colour unless we turn to objects enlightened
by its rays. Moreover it is the Sun which also brings about the
generation, the growth, and the nourishment, of these objects,
though it is itself out of the limits of generation: it generates and
keeps them in existence, besides rendering them visible.195 Now the
Sun is the offspring and representative of the Idea of Good: what
the Sun is in the sensible and visible world, the Idea of Good is in
the intelligible or conceivable world.196 As the Sun not only brings
into being the objects of sense, but imparts to them the power of
being seen so the Idea of Good brings into being the objects of
conception or cognition, imparts to them the power of being known,
and to the mind the power of knowing them.197 It is from the Idea
of Good that all knowledge, all truth, and all real essence spring. Yet
the Idea of Good is itself extra-essential; out of or beyond the limits
of essence, and superior in beauty and dignity both to knowledge
and to truth; which are not Good itself, but akin to Good, as vision is
akin to the Sun.198

192 Plato, Republic, vi. p. 506 B-E. Αὐτὸ μὲν τί ποτ’

ἐστὶ τἀγαθὸν ἐάσωμεν τὰ νῦν εἶναι· πλέον γάρ μοι
φαίνεται ἢ κατὰ τὴν παροῦσαν ὁρμὴν ἐφικέσθαι τοῦ γε
δοκοῦντος ἐμοὶ τὰ νῦν· ὅς δὲ ἔκγονός τε τοῦ ἀγαθοῦ
φαίνεται καὶ ὁμοιότατος ἐκείνῳ, λέγειν ἐθέλω (p. 506
E).

193 Plato, Republic, vi. p. 507 B-C. Καὶ τὰ μὲν (πολλὰ)

δὴ ὁρᾶσθαί φαμεν, νοεῖσθαι δὲ οὔ· τὰς δ’ αὖ ἰδέας
νοεῖσθαι μέν, ὁρᾶσθαι δὲ οὔ.

194 Plato, Republic, vi. p. 508 A. ἡ ὄψις —

ἡλιοειδέστατον τῶν περὶ τὰς αἰσθήσεις ὀργάνων.

195 Plato, Republic, vi. p. 509 B. Τὸν ἥλιον τοῖς

ὁρωμένοις οὐ μόνον τὴν τοῦ ὁρᾶσθαι δύναμιν παρέχειν
φήσεις, ἀλλὰ καὶ τὴν γένεσιν καὶ αὔξην καὶ τροφήν, οὐ
γένεσιν αὐτὸν ὄντα.
 196 Plato, Republic, vi. p. 508 B-C. Τοῦτον (τὸν
ἥλιον) τὸν τοῦ ἀγαθοῦ ἔκγονον, ὃν τἀγαθὸν ἐγέννησεν
ἀνάλογον ἑαυτῷ, ὅ, τι περ αὐτὸ ἐν τῷ νοητῷ τόπῳ
πρός τε νοῦν καὶ τὰ νοούμενα, τοῦτο τοῦτον ἐν τῷ
ὁρατῷ πρός τε ὄψιν καὶ τὰ ὁρώμενα.

197 Plato, Republic, vi. p. 508 E. Τοῦτο τοίνυν τὸ τὴν

ἀλήθειαν παρέχον τοῖς γιγνωσκομένοις καὶ τῷ
γιγνώσκοντι τὴν δύναμιν ἀποδιδὸν τὴν τοῦ ἀγαθοῦ
ἰδέαν φάθι εἶναι, αἰτίαν δ’ ἐπιστήμης οὖσαν καὶ
ἀληθείας ὡς γιγνωσκομένης, &c.

198 Plato, Republic, vi. p. 509 B. Καὶ τοῖς

γιγνωσκομένοις τοίνυν μὴ μόνον τὸ γιγνώσκεσθαι
φάναι ὑπὸ τοῦ ἀγαθοῦ παρεῖναι, ἀλλὰ καὶ τὸ εἶναι τε
καὶ τὴν οὐσίαν ὑπ’ ἐκείνου αὐτοις προσεῖναι, οὐκ
οὐσίας ὄντος τοῦ ἀγαθοῦ, ἀλλ’ ἔτι ἐπέκεινα τῆς οὐσίας
πρεσβείᾳ καὶ δυνάμει ὑπερέχοντος. Καὶ ὁ Γλαύκων
μάλα γελοίως, Ἄπολλον, ἔφη, δαιμονίας ὑπερβολῆς! Σὺ
γάρ, ἦν δ’ ἐγώ, αἴτιος, ἀναγκάζων τὰ ἐμοὶ δοκοῦντα
περὶ αὐτοῦ λέγειν. — Also p. 509 A.

Here then we have two distinct

The Idea of Good rules the
regions or genera; one, the conceivable
ideal or intelligible world,
or intelligible, ruled by the Idea of
as the Sun rules the
Good — the other the visible, ruled by
sensible or visible world.
the Sun, which is the offspring of
Good. Now let us subdivide each of these regions or genera, into
two portions. The two portions of the visible will be — first, real
objects, visible such as animals, plants, works of art, &c. — second,
the images or representations of these, such as shadows, reflexions
in water or in mirrors, &c. The first of these two subdivisions will be
greatly superior in clearness to the second: it will be distinguished
from the second as truth is distinguished from not-truth.199 Matter of
knowledge is in the same relation to matter of opinion, as an original
to its copy. Next, the conceivable or intelligible region must be
subdivided into two portions, similarly related one to the other: the
first of these portions will be analogous to the real objects of vision,
the second to the images or representations of these objects: the
first will thus be the Forms, Ideas, or Realities of Conception or
Intellect — the second will be particular images or embodiments
thereof.200

199 Plato, Republic, vi. pp. 509-510. 510 A: διῃρῆσθαι

ἀληθείᾳ τε καὶ μή, ὡς τὸ δοξαστὸν πρὸς τὸ γνωστόν,
οὔτω τὸ ὁμοιωθὲν πρὸς τὸ ᾧ ὡμοιώθη.

200 Plato, Republic, vi. p. 510 B.

Now in regard to these two portions

To the intelligible world
of the conceivable or intelligible region,
there are applicable two
two different procedures of the mind
distinct modes of
are employed: the pure Dialectic, and
procedure — the
the Geometrical, procedure. The
Geometrical — the
Geometer or the Arithmetician begins
Dialectic. Geometrical
with certain visible images, lines,
procedure assumes
figures, or numbered objects, of sense:
diagrams.
he takes his departure from certain
hypotheses or assumptions, such as given numbers, odd and even
— given figures and angles, of three different sorts.201 He assumes
these as data without rendering account of them, or allowing them
to be called in question, as if they were self-evident to every one.
From these premisses he deduces his conclusions, carrying them
down by uncontradicted steps to the solution of the problem which
he is examining.202 But though he has before his eyes the visible
parallelogram inscribed on the sand, with its visible diagonal, and
though all his propositions are affirmed respecting these — yet what
he has really in his mind is something quite different — the
Parallelogram per se, or the Form of a Parallelogram — the Form of
a Diagonal, &c. The visible figure before him is used only as an
image or representative of this self-existent form; which last he can
contemplate only in conception, though all his propositions are
intended to apply to it.203 He is unable to take his departure directly
from this Form, as from a first principle: he is forced to assume the
visible figure as his point of departure, and cannot ascend above it:
he treats it as something privileged and self-evident.204

201 Plato, Republic, vi. p. 510 B. ᾗ το μὲν αὐτοῦ

(τμῆμα) τοῖς τότε τμηθεῖσιν ὡς εἰκόσι χρωμένη (this is
farther illustrated by p. 511 A — εἰκόσι χρωμένην
αὐτοῖς τοῖς ὑπὸ τῶν κάτω ἀπεικασθεῖσἰ) ψυχὴ ζητεῖν
ἀναγκάζεται ἐξ ὑποθέσεων, οὐκ ἐπ’ ἀρχὴν πορευομένη
ἀλλ’ ἐπὶ τελευτήν, &c.

202 Plato, Republic, vi. p. 510 C-D. οἱ περὶ τὰς

γεωμετρίας τε καὶ λογισμοὺς καὶ τὰ τοιαῦτα
πραγματευόμενοι, ὑποθέμενοι τό τε περιττὸν καὶ τὸ
ἄρτιον καὶ τὰ σχήματα καὶ γωνιῶν τριττὰ εἴδη καὶ ἄλλα
τούτων ἀδελφὰ καθ’ ἑκάστην μέθοδον, ταῦτα μὲν ὡς
εἰδότες, ποιησάμενοι ὑποθέσεις αὐτά, οὐδένα λόγον
οὔτε αὑτοῖς οὔτε τοῖς ἄλλοις ἔτι ἀξιοῦσι περὶ αὐτῶν
διδόναι, ὡς παντὶ φανερῶν· ἐκ τούτων δ’ ἀρχόμενοι τὰ
λοιπὰ ἤδη διεξιόντες τελευτῶσιν ὁμολογουμένως ἐπὶ
τοῦτο, οὖ ἂν ἐπὶ σκέψιν ὁρμήσωσιν.
 203 Plato, Republic, vi. p. 510 D-E. τοῖς ὁρωμένοις
εἴδεσι προσχρῶνται, καὶ τοὺς λόγους περὶ αὐτῶν
ποιοῦνται, οὐ περὶ τούτων διανοούμενοι, ἀλλ’ ἐκείνων
πέρι οἷς ταῦτα ἔοικε, τοῦ τετραγώνου αὐτοῦ ἕνεκα
τοὺς λόγους ποιούμενοι καὶ διαμέτρον αὐτῆς, ἀλλ’ οὐ
ταύτης ἣν γράφουσι, καὶ τἄλλα οὕτως· αὐτὰ μὲν ταῦτα
ἃ πλάττουσί τε καὶ γράφουσιν, ὧν καὶ σκιαὶ καὶ ἐν
ὕδασιν εἰκόνες εἰσί, τούτοις μὲν ὡς εἰκόσιν αὖ
χρώμενοι, ζητοῦντές τε αὐτὰ ἐκεῖνα ἰδεῖν, ἃ οὐκ ἂν
ἄλλως ἴδοι τις ἢ τῇ διανοίᾳ.

204 Plato, Republic, vi. p. 511 A. οὐκ ἐπ’ ἀρχὴν

ἰοῦσαν, ὡς οὐ δυναμένην τῶν ὑποθέσεων ἀνωτέρω
ἐκβαίνειν, εἰκόσι δὲ χρωμένην αὐτοῖς τοῖς ὑπὸ τῶν
κάτω ἀπεικασθεῖσιν, καὶ ἐκείνοις πρὸς ἐκεῖνα ὡς
ἐναργέσι δεδοξασμένοις τε καὶ τετιμημένοις.

From the geometrical procedure thus

Dialectic procedure
described, we must now distinguish the
assumes nothing. It
other section — the pure Dialectic.
departs from the highest
Here the Intellect ascends to the
Form, and steps gradually
absolute Form, and grasps it directly.
down to the lowest,
Particular assumptions or hypotheses
without meddling with any
are indeed employed, but only as
thing except Forms.
intervening stepping-stones, by which
the Intellect is to ascend to the Form: they are afterwards to be
discarded: they are not used here for first principles of reasoning, as
they are by the Geometer.205 The Dialectician uses for his first
principle the highest absolute Form; he descends from this to the
next highest, and so lower and lower through the orderly gradation
of Forms, until he comes to the end or lowest: never employing
throughout the whole descent any hypothesis or assumption, nor
any illustrative aid from sense. He contemplates and reasons upon
the pure intelligible essence, directly and immediately: whereas the
Geometer can only contemplate it indirectly and mediately, through
the intervening aid of particular assumptions.206

205 Plato, Republic, vi. p. 511 B. τὸ ἕτερον τμῆμα τοῦ

νοητοῦ … οὖ αὐτὸς ὁ λόγος ἅπτεται τῇ τοῦ
διαλέγεσθαι δυνάμει, τὰς ὑποθέσεις ποιούμενος οὐκ
ἀρχὰς ἀλλὰ τῷ ὄντι ὑποθέσεις, οἷον ἐπιβάσεις τε καὶ
ὁρμάς, ἵνα μέχρι τοῦ ἀνυποθέτου, ἐπὶ τὴν τοῦ παντὸς
ἀρχὴν ἰών, ἁψάμενος αὐτῆς, πάλιν αὖ ἐχόμενος τῶν
ἐκείνης ἐχομένων, οὕτως ἐπὶ τελευτὴν καταβαίνῃ,
αἰσθητῷ παντάπασιν οὐδενὶ προσχρώμενος, ἀλλ’
εἴδεσιν αὐτοῖς δι’ αὐτῶν εἰς αὐτά, καὶ τελευτᾷ εἰς εἴδη.

206 Plato, Republic, vi. p. 511 C. σαφέστερον εἶναι τὸ

ὑπὸ τῆς τοῦ διαλέγεσθαι ἐπιστήμης τοῦ ὄντος τε καὶ
νοητοῦ θεωρούμενον ἢ τὸ ὑπὸ τῶν τεχνῶν
καλουμένων, αἷς αἱ ὑποθέσεις ἀρχαί, &c.

The distinction here indicated

Two distinct grades of
between the two different sections of
Cognition — Direct or
the Intelligible Region, and the two
Superior — Noûs — Indirect
different sections of the Region of
or Inferior — Dianoia.
Sense — we shall mark (continues
Sokrates) by appropriate terms. The Dialectician alone has Noûs or
Intellect, direct or the highest cognition: he alone grasps and
comprehends directly the pure intelligible essence or absolute Form.
The Geometer does not ascend to this direct contemplation or
intuition of the Form: he knows it only through the medium of
particular assumptions, by indirect Cognition or Dianoia; which is a
lower faculty than Noûs or Intellect, yet nevertheless higher than
Opinion.

As we assign two distinct grades of

Two distinct grades of
Cognition to the Intelligible Region, so
Opinion also in the Sensible
we also assign two distinct grades of
World — Faith or Belief —
Opinion to the Region of Sense, and its
Conjecture.
two sections. To the first of these two
sections, or to real objects of sense, we assign the highest grade of
Opinion, viz.: Faith or Belief. To the second of the two, or to the
images of real objects of sense, we assign the lower grade, viz.:
Conjecture.

Here then are the four grades. Two grades of Cognition — 1.

Noûs, or Direct Cognition. 2. Dianoia, or Indirect Cognition: both of
them belonging to the Intelligible Region, and both of them higher
than Opinion. Next follow the two grades of Opinion. 3. The higher
grade, Faith or Belief. 4. The lower grade, Conjecture. Both the two
last belong to the sensible world; the first to real objects, the last to
images of those objects.207

207 Plato, Republic, p. 511 D-E.

Sokrates now proceeds to illustrate

Distinction between the
the contrast between the philosopher
philosopher and the
and the unphilosophical or ordinary
unphilosophical public,
man, by the memorable simile of the
illustrated by the simile of
cave and its shadows. Mankind live in a
the Cave, and the captives
cave, with its aperture directed towards
imprisoned therein.
the light of the sun; but they are so
chained, that their backs are constantly turned towards this
aperture, so that they cannot see the sun and sunlight. What they
do see is by means of a fire which is always burning behind them.
Between them and this fire there is a wall; along the wall are posted
men who carry backwards and forwards representations or images
of all sorts of objects; so that the shadows of these objects by the
firelight are projected from behind these chained men upon the
ground in front of them, and pass to and fro before their vision. All
the experience which such chained men acquire, consists in what
they observe of the appearance and disappearance, the transition,
sequences, and co-existences, of these shadows, which they mistake
for truth and realities, having no no acquaintance with any other
phenomena.208 If now we suppose any one of them to be liberated
from his chains, turned round, and brought up to the light of the sun
and to real objects — his eyesight would be at first altogether
dazzled, confounded, and distressed. Distinguishing as yet nothing
clearly, he would believe that the shadows which he had seen in his
former state were true and distinct objects, and that the new mode
of vision to which he had been suddenly introduced was illusory and
unprofitable. He would require a long time to accustom him to
daylight: at first his eyes would bear nothing but shadows — next
images in the water — then the stars at night — lastly, the full
brightness of the Sun. He would learn that it was the Sun which not
only gave light, but was the cause of varying seasons, growth, and
all the productions of the visible world. And when his mind had been
thus opened, he would consider himself much to be envied for the
change, looking back with pity on his companions still in the cave.209
He would think them all miserably ignorant, as being conversant not
with realities, but only with the shadows which passed before their
eyes. He would have no esteem even for the chosen few in the cave,
who were honoured by their fellows as having best observed the co-
existences and sequences among these shadows, so as to predict
most exactly how the shadows would appear in future.210 Moreover
if, after having become fully accustomed to daylight and the
contemplation of realities, he were to descend again into the cave,
his eyesight would be dim and confused in that comparative
darkness; so that he would not well recognise the shadows, and
would get into disputes about them with his companions. They on
their side would deride him as having spoilt his sight as well as his
judgment, and would point him out as an example to deter others
from emerging out of the cave into daylight.211 Far from wishing to
emerge themselves, they would kill, if they could, any one who tried
to unchain them and assist them in escaping.212

208 Plato, Republic, vii. pp. 514-515.

209 Plato, Republic, vii. pp. 515-516.

210 Plato, Republic, vii. p. 516 C. Τιμαὶ δὲ καὶ ἔπαινοι

εἴ τινες αὐτοῖς ἦσαν τότε παρ’ ἀλλήλων καὶ γέρα τῷ
ὀξύτατα καθορῶντι τὰ παριόντα, καὶ μνημονεύοντι
μάλιστα ὅσα τε πρότερα αὐτῶν καὶ ὕστερα εἰώθει καὶ
ἅμα πορεύεσθαι, καὶ ἐκ τούτων δὴ δυνατώτατα
ἀπομαντευομένῳ τὸ μέλλον ἥξειν, δοκεῖς ἂν αὐτὸν
ἐπιθυμητικῶς αὐτῶν ἔχειν καὶ ζηλοῦν τοὺς παρ’
ἐκείνοις τιμωμένους τε καὶ ἐνδυναστεύοντας;

211 Plato, Republic, vii. p. 517 A. ἆρ’ οὐ γέλωτ’ ἂν

παράσχοι καὶ λέγοιτο ἂν περὶ αὐτοῦ ὡς ἀναβὰς ἄνω
διεφθαρμένος ἥκει τὰ ὄμματα, καὶ ὅτι οὐκ ἄξιον οὐδὲ
πειρᾶσθαι ἄνω ἰέναι;

212 Plato, Republic, vii. p. 517 A. καὶ τὸν ἐπιχειροῦντα

λύειν τε καὶ ἀνάγειν, εἴ πως ἐν ταῖς χερσὶ δύναιτο
 λαβεῖν καὶ ἀποκτεῖναι, ἀποκτιννύναι ἂν;

By this simile (continues Sokrates) I

Daylight of philosophy
intend to illustrate, as far as I can, yet
contrasted with the
without speaking confidently,213 the
firelight and shadows of
the Cave.
relations of the sensible world to the
intelligible world: the world of
transitory shadows, dimly seen and admitting only opinion,
contrasted with that of unchangeable realities steadily contemplated
and known, illuminated by the Idea of Good, which is itself visible in
the background, being the cause both of truth in speculation and of
rectitude in action.214 No wonder that the few who can ascend into
the intelligible region, amidst the clear contemplations of Truth and
Justice per se, are averse to meddle again with the miseries of
human affairs and to contend with the opinions formed by ordinary
men respecting the shadows of Justice, the reality of which these
ordinary men have never seen. There are two causes of temporary
confused vision: one, when a man moves out of darkness into light
— the other when he moves from light into darkness. It is from the
latter cause that the philosopher suffers when he redescends into
the obscure cave.215

213 Plato, Republic, vii. p. 517. τῆς γ’ ἐμῆς ἐλπίδος,

ἐπειδὴ ταύτης ἐπιθυμεῖς ἀκούειν· θ ε ὸ ς δ έ π ο υ
οἶδεν εἰ ἀληθὴς οὖσα τυγχάνει.

This tone of uncertainty in Plato deserves notice. It

forms a striking contrast with the dogmatism of many
among his commentators.

214 Plato, Republic, vii. p. 517 C.

 215 Plato, Republic, vii. pp. 517-518.

The great purpose of education is to

Purpose of a philosophical
turn a man round from his natural
training, to turn a man
position at the bottom of this dark
round from facing the bad
cave, where he sees nothing but
light of the Cave to face the
shadows: to fix his eyes in the other
daylight of philosophy, and
direction, and to induce him to ascend
to see the eternal Forms.
into clear daylight. Education does not,
as some suppose, either pour knowledge into an empty mind, or
impart visual power to blind persons. Men have good eyes, but these
eyes are turned in the wrong direction. The clever among them see
sharply enough what is before them: but they have nothing before
them except shadows, and the sharper their vision the more
mischief they do.216 What is required is to turn them round and draw
them up so as to face the real objects of daylight. Their natural
eyesight would then suffice to enable them to see these objects
well.217 The task of our education must be, to turn round the men of
superior natural aptitude, and to draw them up into the daylight of
realities. Next, when they shall have become sufficiently initiated in
truth and philosophy, we must not allow them to bury themselves
permanently in such studies — as they will themselves be but too
eager to do. We must compel them to come down again into the
cave and exercise ascendancy among their companions, for whose
benefit their superior mental condition will thus become available.218

216 Plato, Republic, p. 519 A-B.

217 Plato, Republic, p. 519 B. ὧν εἰ ἀπαλλαγὲν

περιεστρέφετο εἰς τἀληθῆ, καὶ ἐκεῖνα ἂν τὸ αὐτὸ τοῦτο
τῶν αὐτῶν ἀνθρώπων ὀξύτατα ἑώρα, ὥσπερ καὶ ἐφ’ ἂ
νῦν τέτραπται.
 218 Plato, Republic, vii. pp. 519-520.

Coming as they do from the better

Those who have emerged
light, they will, after a little temporary
from the Cave into full
perplexity, be able to see the dim
daylight amidst eternal
shadows better than those who have
Forms, must be forced to
never looked at anything else. Having
come down again and
contemplated the true and real Forms
undertake active duties —
of the Just, Beautiful, Good — they will
Their reluctance to do this.
better appreciate the images of these
Forms which come and go, pass by and repass in the cave.219 They
will indeed be very reluctant to undertake the duties or exercise the
powers of government: their genuine delight is in philosophy; and if
left to themselves, they would cultivate nothing else. But such
reluctance is in itself one proof that they are the fittest persons to
govern. If government be placed in the hands of men eager to
possess it, there will be others eager to dispossess them, so that
competition and factions will arise. Those who come forward to
govern, having no good of their own, and seeking to extract their
own good from the exercise of power, are both unworthy of trust
and sure to be resisted by opponents of the like disposition. The
philosopher alone has his own good in himself. He enjoys a life
better than that of a ruler; which life he is compelled to forego when
he accepts power and becomes a ruler.220

219 Plato, Republic, vii. p. 520 C.

220 Plato, Republic, vii. pp. 520-521.

The main purpose of education, I

Studies serving as
have said (continues Sokrates) is, to
introduction to philosophy
turn round the faces of the superior
— Arithmetic, its
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com