Nothing Special   »   [go: up one dir, main page]

PDF Supervised Machine Learning Optimization Framework and Applications With Sas and R First Edition Kolosova Ebook Full Chapter

Download as pdf or txt
Download as pdf or txt
You are on page 1of 54

Supervised machine learning:

optimization framework and


applications with SAS and R First
Edition Kolosova
Visit to download the full and correct content document:
https://textbookfull.com/product/supervised-machine-learning-optimization-framework-
and-applications-with-sas-and-r-first-edition-kolosova/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Machine Learning Foundations: Supervised, Unsupervised,


and Advanced Learning Taeho Jo

https://textbookfull.com/product/machine-learning-foundations-
supervised-unsupervised-and-advanced-learning-taeho-jo/

Fundamentals of optimization theory with applications


to machine learning Gallier J.

https://textbookfull.com/product/fundamentals-of-optimization-
theory-with-applications-to-machine-learning-gallier-j/

Fundamentals of optimization theory with applications


to machine learning Gallier J.

https://textbookfull.com/product/fundamentals-of-optimization-
theory-with-applications-to-machine-learning-gallier-j-2/

Introduction to machine learning with R rigorous


mathematical analysis First Edition Burger

https://textbookfull.com/product/introduction-to-machine-
learning-with-r-rigorous-mathematical-analysis-first-edition-
burger/
Applied Analytics through Case Studies Using SAS and R:
Implementing Predictive Models and Machine Learning
Techniques Deepti Gupta

https://textbookfull.com/product/applied-analytics-through-case-
studies-using-sas-and-r-implementing-predictive-models-and-
machine-learning-techniques-deepti-gupta/

Supervised Learning with Quantum Computers Maria Schuld

https://textbookfull.com/product/supervised-learning-with-
quantum-computers-maria-schuld/

Supervised Machine Learning in Wind Forecasting and


Ramp Event Prediction 1st Edition Harsh S. Dhiman

https://textbookfull.com/product/supervised-machine-learning-in-
wind-forecasting-and-ramp-event-prediction-1st-edition-harsh-s-
dhiman/

Blockchain, big data and machine learning: trends and


applications First Edition Kumar

https://textbookfull.com/product/blockchain-big-data-and-machine-
learning-trends-and-applications-first-edition-kumar/

Teaching Learning Based Optimization Algorithm And Its


Engineering Applications 1st Edition R. Venkata Rao

https://textbookfull.com/product/teaching-learning-based-
optimization-algorithm-and-its-engineering-applications-1st-
edition-r-venkata-rao/
Supervised Machine
Learning
Supervised Machine
Learning
Optimization Framework and
Applications with SAS and R

Tanya Kolosova PhD


Associates in Analytics Inc., Boca Raton, Florida
Samuel Berestizhevsky MSc
Associates in Analytics Inc., Boca Raton, Florida
First edition published 2021
by CRC Press
6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742

and by CRC Press


2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

© 2021 Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, LLC

This book contains information obtained from authentic and highly regarded sources. While all
reasonable efforts have been made to publish reliable data and information, neither the author[s]
nor the publisher can accept any legal responsibility or liability for any errors or omissions that may
be made. The publishers wish to make clear that any views or opinions expressed in this book by
individual editors, authors or contributors are personal to them and do not necessarily reflect the
views/opinions of the publishers. The information or guidance contained in this book is intended
for use by medical, scientific or health-care professionals and is provided strictly as a supplement to
the medical or other professional’s own judgement, their knowledge of the patient’s medical history,
relevant manufacturer’s instructions and the appropriate best practice guidelines. Because of the
rapid advances in medical science, any information or advice on dosages, procedures or diagnoses
should be independently verified. The reader is strongly urged to consult the relevant national drug
formulary and the drug companies’ and device or material manufacturers’ printed instructions, and
their websites, before administering or utilizing any of the drugs, devices or materials mentioned in
this book. This book does not indicate whether a particular treatment is appropriate or suitable for
a particular individual. Ultimately it is the sole responsibility of the medical professional to make
his or her own professional judgements, so as to advise and treat patients appropriately. The authors
and publishers have also attempted to trace the copyright holders of all material reproduced in this
publication and apologize to copyright holders if permission to publish in this form has not been
obtained. If any copyright material has not been acknowledged please write and let us know so we
may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.
com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA
01923, 978-750-8400. For works that are not available on CCC please contact mpkbookspermissions@
tandf.co.uk

Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.

ISBN: 978-0-367-27732-1 (hbk)


ISBN: 978-0-367-53882-8 (pbk)
ISBN: 978-0-429-29759-5 (ebk)

Typeset in Palatino
by codeMantra
Dedication

To our children—love and meaning of our lives.


Contents

Acknowledgments .............................................................................................. xiii


Authors ...................................................................................................................xv
Introduction: Challenges in the Application of Machine
Learning Classification Methods ..................................................................... xvii

Part I
1. Introduction to the AI Framework .............................................................. 3
1.1 Components of the AI Framework and Their Interaction...............3
1.2 AI Framework in Detail ....................................................................... 5
1.2.1 Creating Training and Test Datasets..................................... 5
1.2.2 Design of Experiments for a Classifier ................................. 6
1.2.3 Firth Logistic Regression ........................................................6
1.2.4 Data
 Contamination ................................................................6
1.2.5 Best Classifiers .......................................................................... 7
1.3 SAS Procedures for the AI Framework Components ...................... 7
1.4 R Libraries for the AI Framework Components ............................... 7
References .........................................................................................................8

2. Supervised Machine Learning and Its Deployment in SAS and R .....9


2.1 Introduction ...........................................................................................9
2.2 Principles of Supervised Machine Learning .................................. 10
2.3 Neural Network .................................................................................. 12
2.3.1 Introduction ............................................................................ 12
2.3.2 Neural Network Components.............................................. 13
2.3.2.1 Activation Function ............................................... 13
2.3.2.2 Neurons ................................................................... 14
2.3.2.3 Networks ................................................................. 15
2.3.3 R for Neural Networks .......................................................... 16
2.4 Support Vector Machine .................................................................... 16
2.4.1 Introduction
 ............................................................................ 16
2.4.2 Kernel....................................................................................... 18
2.4.3 Margin ..................................................................................... 19
2.4.4 Optimization .......................................................................... 20
2.4.5 Bias–Variance Trade-off and SVM Hyperparameters ...... 20
2.4.6 R for SVM ................................................................................ 21
2.5 SVM Modification Using Firth’s Regression ...................................22
2.5.1 Introduction
 ............................................................................22
2.5.2 Logistic Regression ................................................................ 23
2.5.3 Problem of Separation ........................................................... 23

vii
viii Contents

2.5.4 R for Firth’s Regression ......................................................... 25


2.5.5 SAS for Firth’s Regression..................................................... 25
2.6 Summary
 .............................................................................................. 27
References ....................................................................................................... 27

3. Bootstrap Methods and Their Deployment in SAS and R................... 29


3.1 Introduction ......................................................................................... 29
3.2 Overview of Bootstrap Methods.......................................................30
3.2.1 The Basic Bootstrap ............................................................... 31
3.2.2 Hypothesis Tests, Estimates, and Confidence Intervals... 32
3.2.3 Bias Reduction ........................................................................34
3.2.4 The Parametric Bootstrap ..................................................... 35
3.2.5 m-out-of-n Bootstrap ............................................................. 36
3.2.6 Bootstrap Samples Similarity ............................................... 36
3.3 Implementation of Bootstrap in SAS and R .................................... 37
3.3.1 m-out-of-n in SAS................................................................... 37
3.3.2 m-out-of-n in R ....................................................................... 38
3.4 Summary
 .............................................................................................. 39
References ....................................................................................................... 40

4. Outliers Detection and Its Deployment in SAS and R ......................... 41


4.1 Introduction
 ......................................................................................... 41
4.2 Outliers Detection and Treatment ....................................................42
4.2.1 Minimum Covariance Determinant Method ....................42
4.2.2 MCD in SAS ............................................................................43
4.3 Bias Reduction .....................................................................................44
4.4 Summary .............................................................................................. 45
References ....................................................................................................... 45

5. Design of Experiments and Its Deployment in SAS and R ................. 47


5.1 Introduction
 ......................................................................................... 47
5.2 Application of DoE in AI Framework .............................................. 48
5.2.1 Terminology of DoE .............................................................. 49
5.2.1.1 Experiment .............................................................. 49
5.2.1.2 Experimental Unit .................................................. 49
5.2.1.3 Factor ........................................................................ 49
5.2.1.4 Treatment ................................................................. 49
5.2.2 Principles of DoE.................................................................... 49
5.2.2.1 Randomization ....................................................... 49
5.2.2.2 Statistical Replication............................................. 50
5.2.2.3 Blocking ................................................................... 50
5.2.2.4 Orthogonality ......................................................... 50
5.2.3 Full-Factorial
 Experiment ..................................................... 50
5.2.4 Fractional Factorial Experiment .......................................... 57
5.2.5 Linear Mixed Models ............................................................ 58
Contents ix

5.2.6 Factors and Response Variables in the AI Framework ..... 59


5.2.7 Example ................................................................................... 60
5.2.8 Analysis of Linear Mixed Model Using SAS ..................... 62
5.2.9 Analysis of Linear Mixed Model Using R ..........................65
5.3 Summary .............................................................................................. 66
References ....................................................................................................... 67

Part II

6.1 Principles of Code-Free Design......................................................... 71


6.2 The Data Dictionary Components for the AI Framework ............ 72
6.2.1 Relational Model .................................................................... 72
6.2.2 Table ......................................................................................... 73
6.2.3 Data Aspects ........................................................................... 73
6.2.4 Relational Data Structure ..................................................... 73
6.2.5 Domains .................................................................................. 74
6.2.6 Relations and Tables .............................................................. 74
6.2.7 Functions ................................................................................. 74
6.2.8 One-to-one Relationship ....................................................... 75
6.2.9 One-to-many Relationship ................................................... 75
6.2.10 Primary Key............................................................................ 75
6.2.11 Foreign Key ............................................................................. 75
6.2.12 Missing Values ....................................................................... 75
6.2.13 Data Dictionary ...................................................................... 76
6.3 Properties of the Data Dictionary ..................................................... 76
6.3.1 The Library Table ................................................................... 76
6.3.2 The Object Table .....................................................................77
6.3.3 The Location Table .................................................................77
6.3.4 The Message Table .................................................................77
6.3.5 The Property Table ................................................................ 78
6.3.6 Meaning .................................................................................. 79
6.3.7 The Link Table ........................................................................ 79
6.3.8 Process of Application Data Model Definition .................. 79
6.3.9 Features of the Data Dictionary ...........................................80
6.3.10 The Components of the Optimization Framework
and Their Definitions in the Data Dictionary.................... 81
6.4 Deployment of Code-Free Design with SAS and R........................ 81
6.4.1 How to Generate Application Objects ................................ 81
6.4.2 Generating R Datasets from the Data
Dictionary Metadata..............................................................84
6.4.3 SAS and R Interoperability ................................................... 86
6.5 Summary .............................................................................................. 88
Reference ......................................................................................................... 88
x Contents

7. Input Data Component ................................................................................ 89


7.1 Overview of Data Management ........................................................ 89
7.1.1 Data Dictionary ...................................................................... 89
7.1.1.1 The Input Data Dictionary .................................... 89
7.1.1.2 Input and Structure Tables ...................................90
7.1.1.3 Outlier_Detection and Bias_Correction Tables... 91
7.1.1.4 Bootstrap Table ....................................................... 92
7.1.1.5 Output Table ........................................................... 93
7.1.2 SAS Macro Program .............................................................. 95
7.1.3 R Program ............................................................................... 98
7.2 Summary .............................................................................................. 98

8.1 Data Dictionary ................................................................................... 99


8.1.1 Experiment Table ................................................................. 100
8.1.2 Features Table ....................................................................... 100
8.1.3 Metrics Table ......................................................................... 101
8.1.4 ML_Method Table................................................................ 102
8.1.5 Hyperparameters_Domain Table ...................................... 102
8.1.6 Results Table ......................................................................... 102
8.1.7 Results_Metrics Table .......................................................... 103
8.2 SAS Macro Program ......................................................................... 105
8.3 R  Programs ........................................................................................ 105
8.4 Summary ............................................................................................ 106
Reference ....................................................................................................... 106

9. “Contaminated” Training Datasets Component.................................. 107


9.1 Data Dictionary ................................................................................. 107
9.1.1 Contamination Table ........................................................... 108
9.1.2 Cont_Experiment Table....................................................... 109
9.1.3 Cont_Results Table............................................................... 109
9.1.4 Cont_Metric
 Table ................................................................ 110
9.2 SAS Macro Program ......................................................................... 110
9.3 R  Programs ........................................................................................ 110
9.4 Summary ............................................................................................ 111
Reference ....................................................................................................... 111

Part III

10.1 Introduction
 ....................................................................................... 115
10.2 Review of Underwriters’ Performance .......................................... 116
10.2.1 Metrics of Underwriters’ Performance ............................. 116
10.2.1.1 Hit Ratio ................................................................. 116
Contents xi

10.2.1.2 Conversion
 Rate .................................................... 116
10.2.1.3 Dynamic Conversion Rate .................................. 117
10.2.1.4 Time-to-Deal ......................................................... 118
10.2.2 Analysis of Underwriters’ Performance ........................... 119
10.2.2.1 Data Description................................................... 119
10.2.2.2 Application Flow .................................................. 119
10.2.2.3 Dynamic Conversion Rate per Underwriter .... 121
10.2.2.4 Time-to-Deal per Underwriter ........................... 122
10.3 Traditional Approach to Knowledge Delivery ............................. 123
10.4 Anatomy of Artificial Intelligence Solution .................................. 124

10.4.1 Data Structure ...................................................................... 124
10.4.2 Classification Approach ...................................................... 125
10.4.3 Bias–Variance Trade-Off and SVM Hyperparameters ... 125
10.4.4 Building the Classifier ......................................................... 127
10.4.5 “Contamination” of Training Datasets ............................. 130
10.4.6 Experimental Results .......................................................... 130
10.5 Summary ............................................................................................ 132
References ..................................................................................................... 132

11.1 
Introduction ....................................................................................... 135
11.2 Data ..................................................................................................... 136
11.3 The Cox Model for Claims Event Analysis ................................... 136
11.4 Application of the Cox Model for Claims Analysis ..................... 138
11.4.1 Data Transformation ........................................................... 139
11.4.2 Cox Model Assumption Validation ................................... 141
11.4.3 Bayesian Machine Learning Approach ............................ 144
11.4.4 Deployment with SAS ......................................................... 144
11.4.5 Interpretation of Results ..................................................... 146
11.5 Summary ............................................................................................ 152
References ..................................................................................................... 153

Index ..................................................................................................................... 155


Acknowledgments

Writing this book was a real adventure. It summarizes our experience over
many years of solving problems related to applying machine learning meth-
ods to real-life data. During these years, we met many people who, by sharing
their experience, asking questions, and challenging our solutions, helped to
crystalize this book’s ideas. Among them, Samuel would like to thank Maurice
“Hank” Greenberg for being a unique inspiration and Charles Dangelo for
his expert advice in the insurance domain. Tanya is grateful for the knowl-
edge of the design of experiments shared by Prof. David M. Steinberg from Tel
Aviv University, Israel, and Prof. Carl Schwarz from Simon Fraser University,
Canada. We want to thank the SAS Institute as our long-term software of
choice, as well as the R Core Team and contributors.
We would also like to thank everyone on the Taylor and Francis/CRC
Press team for their support and assistance in preparing this book. Special
thanks to David Grubbs, the acquisition editor, for his great support. Thanks
to Rebecca, Sofia, Varun, and other editorial staff members for helping this
book come to fruition.
Special thanks to our daughter Efrat for her performances of Chopin
and Debussy’s piano works that provided a calming and inspirational
atmosphere.

xiii
Authors

Tanya Kolosova is a statistician, software engineer, educator, and co-author


of two books on statistical analysis and metadata-based applications devel-
opment using SAS. She is an actionable analytics expert and has extensive
knowledge of software development methods and technologies, artificial
intelligence methods and algorithms, and statistically designed experiments.

Samuel Berestizhevsky is a statistician, researcher, and software engineer.


Together with Tanya, Samuel co-authored two books on statistical analy-
sis and metadata-based applications development using SAS. Samuel is an
innovator and expert in the area of automated actionable analytics and arti-
ficial intelligence solutions. His extensive knowledge of software develop-
ment methods, technologies, and algorithms allows him to develop solutions
on the cutting edge of science.

xv
Introduction: Challenges in the
Application of Machine Learning
Classification Methods

According to Bishop (2006), machine learning (ML) is a scientific study of


statistical models and algorithms to help a computing system accomplish
a specific task without using explicit instructions, but relying on patterns
and inference instead. Such patterns and inferences can be extracted from
sample data, also called “training datasets” in the machine learning domain,
using machine learning algorithms.
ML is a subset of artificial intelligence technology. It automatically learns
and improves the performance with the pace of time, interactions, and expe-
riences. Building initial machine learning models, or classifiers, is an itera-
tive process, where we start with initial hypotheses about which data can be
useful and how to structure different sets of features, which machine learn-
ing methods can learn better from our data, and how to “tune” multiple
hyperparameters to achieve robust classification results. The machine learn-
ing development process becomes a nightmare if it does not use a framework
for easy, reliable, meaningful, verifiable, and reproducible development.
A large amount of available ML methods creates a challenge for research-
ers: how to develop, verify, reproduce, and compare models using different
ML methods? Utilizing the same or similar data for modeling by different
methods naturally leads to outcomes that differ in their form or quality or
both. We want to be able to compare results from different models and to
draw conclusions about the utility of using one model over the other. The
design of statistical experiments may work as a framework that helps us to
systematically compare different models. Such experiments, which include
determining which class of models to use and what types of features to
include, produce a number of different results.
Reproducibility is the ability to reproduce the analyses.
Comparability is the ability to compare results produced by different mod-
eling approaches.
In this book, we describe an AI framework that helps developers to achieve
reproducibility and comparability of machine learning experiments and to
define ML development processes in precise, transparent, easily changeable,
and verifiable ways.

xvii
xviii Introduction

Fundamentals of Supervised Classification


Machine learning can be roughly categorized into three types: supervised
learning, unsupervised learning, and reinforcement learning. In this book,
we only deal with supervised machine learning. The most common super-
vised machine learning task is classification.
Supervised classification can be described as the task of automatically
assigning objects to their respective classes on the basis of numerical mea-
surements derived from these objects. Classifiers are the tools that imple-
ment the actual classification from these measurements to the so-called class
labels. The field of supervised classification studies ways of constructing
such classifiers. The main idea behind supervised learning methods is that
of learning from examples: given several so-called training datasets that
describe input–output relations, to what extent can the general mapping be
learned that takes any new and unseen objects to their correct classes?
The basic problem of machine learning is probably the following. We have
a training dataset of cases

( x1 , y1 ) , …, ( xn , yn )
where each case ( xi , y i ) , i = 1, … , n consists of an object xi (a vector of fea-
tures) and its label y i. The problem is to predict the label y n+1 of a new object
xn+1 . Usually, the goal of classification is to produce a prediction ŷ n+1 that is
likely to coincide with the true label y n+1, and this goal should be comple-
mented with some measure of its reliability. There is a clear tradeoff between
accuracy and reliability: we can improve the former by relaxing the latter
and vice versa.

Generalization Problem: Bias–Variance Tradeoff


The quality of a classifier is defined by its ability to correctly classify an input
vector into an appropriate class and is measured by metrics derived from
confusion, or misclassification, matrix. There is no such metric that can mea-
sure a classifier quality alone: at least a pair of metrics should be used. For
example, sensitivity and specificity estimate true positive and true negative
rates, respectively, and we are interested in classifiers with both of these met-
rics approaching 1.
The bias–variance tradeoff is the idea that finding a minimum-risk classi-
fication method involves striking a balance between minimizing bias (being
right on average) and minimizing variance (being stable with respect to
Introduction xix

variation in training datasets). A classification method that performs poorly


due to high variance is said to overfit when presented with data. One that
performs poorly due to high bias is said to underfit.
Low bias means

• the average accuracy metric of classification results is close to the


truth.

High bias means

• the average accuracy metric of classification results is far from the


truth,
• the classification method is not sufficiently flexible, and
• individual results of classification accuracy metric are not adequately
adapted to the data.

Low variance means

• each individual accuracy metric of classification is close to the aver-


age accuracy metric,
• individual accuracy metric tends to be similar to one another, and
• the classification method is “stable” with respect to which dataset
(of all possible datasets) it is applied.

High variance means

• individual accuracy metrics are often far from the average accuracy,
• individual accuracy metrics are quite different from one another,
and
• the classification method is very sensitive to which dataset it is
applied.

The challenge of bias and variance estimation can actually be easily


addressed by applying two readily available statistical methodologies:

• bootstrap and
• mixture experiments.

We demonstrate in this book how the application of m-out-of-n bootstrap


allows creation of thousands of training datasets and in turn how the appli-
cation of the ML method to these datasets allows us to estimate bias and
variance. The statistical approach of the mixture experiment is applied to
xx Introduction

contaminate data and to find the level of contamination that “breaks” the
classifier. This helps to identify robust classifiers that are better positioned
for unseen data.

Challenges of Deployment
The useful life of the machine learning application begins when it is deployed
to production. From this point, application modifications are often required.
These modifications may consume a great deal of effort in recoding and
even redesigning. Generally, the better the machine learning application, the
longer its life. Modifications and maintenance of the application are usually
performed by people other than the original implementers. This problem
is very difficult to solve because the application internals cannot be eas-
ily understood. The code-free framework described in this book addresses
these problems. Generally, the framework outlined in this book addresses
the most fundamental problems of existing machine learning application
development and deployment technologies: the failure to recognize users’
true needs and the inability to develop machine learning solutions quickly
to meet these needs.
Machine learning applications have all the challenges of the traditional
software systems, plus an additional set of machine learning–specific issues –
reproducibility and comparability. These two machine learning–specific
issues can be solved by applying the principles of statistically designed
experiments combined with bootstrap along with testing the robustness
of the machine learning solution through data contamination processes
(flipping or corruption of labels of training dataset). The principles and
applications of designed experiments, bootstrap, and data contamination are
discussed in this book.
Another challenge is the selection of the production classifier. If, for exam-
ple, the criterion of a classifier quality is to maximize sensitivity and specific-
ity, then we may encounter a difficult choice. Let us say, we have classifier A
with a sensitivity of 0.85 and a specificity of 0.88, and classifier B with a sensi-
tivity of 0.87 and a specificity of 0.87 – then which of them should we choose?
In this book, we want to explore a different approach, wherein we do not
need to limit our choice by one classifier. Both classifiers, A and B, have a
very good classification quality. More than that, we can find a tens of clas-
sifiers with very similar sensitivity and specificity. We also know that the
sensitivity and specificity of each classifier will vary depending on the data-
sets they are applied to. So, instead of choosing one classifier, we will use all
classifiers that have high-quality metrics, and the final decision on the classi-
fication result will be based on a voting mechanism using some information
criteria. This approach will immediately improve the classification process.
Introduction xxi

Source Code
The source code for the book can be continuously improved and extended
after the book has been published. The source code is located on GitHub:
https://github.com/smlof/Supervised-Machine-Learning--Optimization-
Framework
Part I
1
Introduction to the AI Framework

The artificial intelligence (AI) framework should be algorithm and tech-


nology agnostic. This means that it should not focus on a specific machine
learning methodology and specific systems or software, but it applies to the
design, application, and use of AI in general.
In this book, we describe the AI framework that helps to create machine
learning classifiers in such a way that they produce a sufficiently good clas-
sification of unseen data. The framework incorporates data dictionaries that
define processes such as statistically designed experiments, bootstrapping,
and data “contamination.” This framework is fully transparent as all data
processes are defined as metadata that ensures comparability and reproduc-
ibility of the results of the supervised classification.
Because the AI framework is fully independent of technology, it can be
implemented using different software platforms. In this book, we present
the implementation of the AI framework using two different software tools:
SAS and R.
SAS is an integrated software system that enables accessing data across
multiple sources, manipulating data, performing sophisticated analyses, and
much more, and is one of the major choices of corporate business, though a
bit costly.
R (R Core Team, 2018) is an open-source programming language devel-
oped by researchers, and many latest statistical and analytical methods are
developed with R. This language has extensive documentation along with
a supportive community and is widely used among statisticians and data
scientists for developing statistical software and data analysis.

1.1 Components of the AI Framework and Their Interaction


The main idea of the AI framework proposed in this book is to create a sys-
tematic approach that resolves the challenges of machine learning modeling.
The process of building a machine learning classifier requires making
choices that very often cannot be informed by intuition, understanding of

3
4 Supervised Machine Learning

a business problem, or pre-analysis of data. For example, making a decision


about the feature subset requires considering different hypotheses, and it is
reasonable to consider several different feature sets. Which one of them to
choose eventually requires to go through the modeling process and com-
parison of classifiers built on different feature sets.
Choosing a machine learning method is challenging as well, and in many
instances, the decision is made based on the convenience of the method and
expected interpretability of the results, rather than based on theoretical and
technical aspects of machine learning methods. If we do not want to limit
ourselves to only one machine learning method, but rather want to try a few
of them, we need to estimate and compare the classifiers built by those meth-
ods. Now we have to keep in mind that the quality of an estimated classifier
depends not only on the machine learning method used, but on the feature
set as well.
For each machine learning method, we need to assign values to the hyper-
parameters of that method, and it is obvious that the values of the hyperpa-
rameters impact the quality of the estimated classifier. Different methods
have different numbers of hyperparameters, and finding an optimal set of
values becomes a very difficult problem.
Figure 1.1 describes some most important components of the AI frame-
work regarding which a decision should be made, but how? The statistical
design of an experiment is a methodology that helps here. It allows us to cre-
ate a plan, execute it, compare the results according to the selected metrics of
the classifier quality, and repeat if needed. It often happens that we end up
with several classifiers that have similar classification quality.

FIGURE 1.1
Components of the AI framework.
Introduction to the AI Framework 5

1.2 AI Framework in Detail


AI framework addresses the challenges of building classifiers. These chal-
lenges include not only the difficulties of making selection of feature sets,
machine learning methods, and values of their hyperparameters based on
the classifier quality, but also the problems that arise during the process of
classifier building. These details are presented in Figure 1.2.

1.2.1 Creating Training and Test Datasets


As a first step, the AI framework deals with creating training and test data-
sets. The familiar approach of dividing available data into two parts, and
being concerned only with a proportion of such a division, e.g., 50-50 and
70-30, has a built-in drawback. It relates to bias and dependency within data.
It also relates to the situation that estimating a classifier on one dataset does
not allow us to measure variance and bias of the classifier quality and does
not allow us to evaluate the classifier robustness. In the developed AI frame-
work, the bootstrap approach is used to address this drawback.

• Bootstrap is used to create multiple—and if data permits, hundreds


or even thousands of—training and test datasets. It allows us to esti-
mate and correct bias in data, increase data variability, and decrease
dependency within data.

FIGURE 1.2
AI framework process.
6 Supervised Machine Learning

• Estimation of a classifier on multiple training datasets produces a


sample of the classifier quality metrics that now allows for a compre-
hensive estimation of the classifier quality.
• Use of multiple, instead of a single, test datasets helps to validate the
ability of the classifier to deal with unseen data.

1.2.2 Design of Experiments for a Classifier


As a next step, the AI framework addresses the question of parameters
­
selection, where the plan is created based on the statistical methodology of
the design of experiments. The plan creates combinations of all parameters of
interest in a way that the results can be analyzed and significant insights can
be drawn. For example, we may consider five possible feature sets, the sup-
port vector machine method with three different kernels, and three intervals
of values for each hyperparameter. The plan (design of experiments) will
define how many different setups should be used to build a classifier and
how many training datasets each setup should be applied to.

1.2.3 Firth Logistic Regression


Classifier results, by the nature of machine learning algorithms, do not have
probabilistic characteristics. This means that if an outcome is predicted as 1,
this prediction is not accompanied by a probability of it being correct. Being
able to associate probabilities with predictions increases the quality of a clas-
sifier. Firth logistic regression is a statistical method that works very well
with classifier prediction, adding to it a probability value.

1.2.4 Data Contamination


One of the most challenging problems of machine learning modeling is
generalization. Building classifiers on multiple training datasets helps to
address bias–variance trade-off, but it still cannot estimate the robustness
of the classifier toward unseen data. Planned contamination of data means
random corruption of labels according to different proportions. For example,
in the case of a binary classifier with labels Yes and No, planned contamina-
tion may look like the following:

1. “Flip” 0.5% of Yes labels to No, and 0.5% of No labels to Yes,


2. “Flip” 0.5% of Yes labels to No, and 1% of No labels to Yes,
3. “Flip” 1% of Yes labels to No, and 0.5% of No labels to Yes,
4. Etc.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Platoons
Phaidoon
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: Platoons Phaidoon


Uit het Grieksch overgebracht

Author: Plato

Translator: P. C. Boutens

Release date: August 26, 2023 [eBook #71489]

Language: Dutch

Original publication: Amsterdam: Maatschappij voor Goede en


Goedkoope Lectuur, 1919

Credits: Wouter Franssen and the Online Distributed Proofreading


Team at https://www.pgdp.net (Koninklijke Bibliotheek, The
Hague)

*** START OF THE PROJECT GUTENBERG EBOOK PLATOONS


PHAIDOON ***
PLATOONS PHAIDOON

WERELD BIBLIOTHEEK

Onder leiding van L. Simons.

UITGEGEVEN DOOR:
DE MAATSCHAPPIJ VOOR GOEDE EN GOEDKOOPE LECTUUR ·
AMSTERDAM

PLATOONS PHAIDOON
UIT HET GRIEKSCH OVERGEBRACHT
DOOR

P. C. BOUTENS

1919
GEDRUKT TER DRUKKERIJ „DE DEGEL” — AMSTERDAM
PERSONEN
Echekrates
Phaidoon
Apollodoros
Sokrates
Kebes
Simmias
Kritoon
Dienaar der Elfmannen.
PLATOONS PHAIDOON
57 Echekrates: Zijt gij zelf, o Phaidoon, bij Sokrates geweest
dien dag waarop hij het gif dronk in de gevangenis? Of hebt gij het
van iemand anders gehoord?
Phaidoon: Ik ben er zelf bij geweest, o Echekrates.
Echekrates: Welke dingen heeft de man dan wel gezegd vóor
zijn dood? En hoe was zijn sterven? Mij zoû het een genot zijn dit te
vernemen. Want het komt tegenwoordig niet veel voor, dat een
mijner medeburgers uit Phlioes te Athenai zich ophoudt, en ook is
B langen-tijd-aan-éen geen gastvriend vandaar aangekomen, die
ons iets zekers omtrent deze dingen zoû hebben kunnen berichten,
—behalve dan dat hij gif gedronken had en zoo gestorven was. Maar
verdere bizonderheden kon hij ons niet geven.
58 Phaidoon: Hebt gij dan ook niet van het geding gehoord, hoe
het daarmeê geloopen is?
Echekrates: Zeker, dat heeft iemand ons bericht. En, o ja, wij
verwonderden ons, dat hij veel later, lang nadat het geding
afgeloopen was, gestorven blijkt te zijn. Hoe kwam dat, Phaidoon?
Phaidoon: Een uitzonderlijk toeval, o Echekrates. Juist toch op
den dag vóor het proces was de achtersteven bekranst van het schip
dat de Atheners naar Delos zenden.
Echekrates: Wat is daarmeê?
Phaidoon: Dat is het schip waarop, naar de Atheners zeggen,
B Theseus indertijd naar Kreta die bekende zeven paren wegvoerde
en weêr behouden mede-terug-bracht. Aan Apolloon nu deden zij,
zooals verhaald wordt, toen de gelofte om, als zij behouden zouden
terugkeeren, elk jaar uit dankbaarheid een feestgezantschap naar
Delos te zenden, dat de Atheners van-toen-af ook nu nog altijd
jaarlijks den god plechtig sturen. Wanneer zij nu een aanvang
gemaakt hebben met de feestzending, moet, naar een wet van hen,
de stad in dien tijd rein zijn en mag men niemand van-staats-wegen
ter-dood-brengen, vóordat het schip in Delos en weêr terug is
aangekomen. Daarmeê gaat nu-en-dan een lange tijd heen,
C wanneer winden hen bij-geval van de vaart houden. De aanvang
der feestzending is het oogenblik wanneer de priester van Apolloon
den achtersteven van het schip bekranst heeft. En dat, zooals ik zeg,
was juist op den dag vóor het rechtsgeding geschied. Daarom heeft
Sokrates ook langen tijd in de gevangenis doorgebracht, tusschen
geding en terechtstelling.
Echekrates: En hoe is het bij zijn dood zelf toegegaan, o
Phaidoon? Wat is er gesproken en gebeurd, en wie van ’s mans
bekenden waren tegenwoordig? Of lieten de bewindvoerders hun
niet toe erbij te zijn, maar is hij eenzaam van vrienden gestorven?
D Phaidoon: Geenerwijs. Er waren erbij tegenwoordig, en wel
velen.
Echekrates: Wees dan zoo vriendelijk ons van dat alles zoo
nauwkeurig mogelijk verslag te geven, of gij moest eenige andere
bezigheid hebben.
Phaidoon: Wel ik hèb den tijd, en zal beproeven het voor u te
doorloopen. Want om mij Sokrates te herinneren, zoowel als ik zelf
over hem spreek als wanneer ik dat een ander hoor doen, is mij
steeds van alles aangenaamst.
Echekrates: Gij kunt ervan op-aan, Phaidoon, dat gij ook zulke
luisteraars zult hebben. Doch tracht zoo nauwkeurig als gij kunt,
alles uiteen te zetten.
E Phaidoon: Ziet, ik-voor-mij heb onder het bijwonen eene
wonderlijke ervaring gehad. Want aan den éenen kant kwam geen
gevoel van medelijden bij mij op als men verwachten zoû bij den
dood van een bevriend man. Want gelukzalig kwam mij de man voor,
o Echekrates, zoowel van gedraging als van woorden: hoe
onbevreesd en eêlgemoed hij eindigde. Zoodat ik den indruk kreeg,
dat hij óok nu hij naar het huis van Hades ging, niet ging buiten
goddelijke beschikking, maar dat hij ook wanneer hij daar gekomen
59zoû zijn, het goed zoû hebben zoo nog ooit eenig ander. Daarom
had ik heelemaal geen gevoel van deernis als natuurlijk zoû lijken in
den bijwoner van een rouwtooneel. Maar anderszins ontbrak ook de
genieting van bezig te zijn met wijsbegeerte gelijk wij die plachten te
hebben—de gesprekken toch waren van die soort—, maar een
eenvoudigweg onverbeeldbaar gevoel was mij bij en een ongewone
menging uit genot tegelijk samengesteld, en uit smart, zoo vaak ik
bedacht dat hij op het oogenblik zoû moeten sterven. En alle
B aanwezigen waren wij vrijwel in dien toestand, nu-eens lachend,
dan-weêr weenend, en éen van ons zelfs onderscheidenlijk,
Apollodoros. Gij kent den man immers wel en zijn aard?
Echekrates: Hoe zoû ik niet?
Phaidoon: Hij dan nu was geheel in dien staat, en ook ik zelf was
geschokt en de overigen.
Echekrates: En wie waren zoo al tegenwoordig, Phaidoon?
Phaidoon: Deze dan, Apollodoros, was erbij van de menschen uit
de stad, en Kritoboelos en zijn vader, en verder Hermogenes en
Epigenes en Aischines en Antisthenes. En ook was er Ktesippos uit
den demos Paiania en Menexenos en eenige andere der
landgenooten. Platoon, meen ik, was ziek.
C Echekrates: En waren er sommigen van buiten?
Phaidoon: Ja, Simmias uit Thebai en Kebes en Phaidoondes, en
uit Megara Eukleides en Terpsioon.
Echekrates: Hoe nu? Waren Aristippos en Kleombrotos niet
tegenwoordig?
Phaidoon: Neen, bepaald niet. Want men zeî nog, dat zij op
Aigina waren.
Echekrates: Was er nog iemand anders?
Phaidoon: Ik denk dat dat vrijwel zijn wie tegenwoordig waren.
Echekrates: Hoe dan verder? Welke, zegt gij, waren de
gesprekken?
D Phaidoon: Ik zal alles met u van het begin af trachten door te
gaan. Voortdurend namelijk, ook de voorafgaande dagen, waren wij
gewoon Sokrates te bezoeken zoowel ik als de overigen. Wij
kwamen dan ’s morgens vroeg bijeen in de gerechtszaal waar ook
het geding plaats gehad had; want die was vlak bij de gevangenis.
Wij wachtten daar dan telkens tot de gevangenis geopend werd, en
onderhielden ons met elkander; want zij ging niet vroeg open. En
wanneer opengedaan was, gingen wij naar binnen tot Sokrates en
brachten meest den heelen dag met hem door. Ook waren wij dien
dag vroeger samengekomen. Want den vorigen dag toen wij ’s
E avonds uit de gevangenis gekomen waren, hadden wij vernomen
dat het schip uit Delos aangekomen was. Wij spraken derhalve met
elkander af om zoo vroeg mogelijk naar de gewone plaats te komen.
Toen wij er dan waren, kwam de deurwachter die ons placht open te
doen, naar buiten en zeide ons te wachten en niet eerder te komen
aankloppen, totdat hij zelf ons binnennooden zoû. Want, zeî hij, de
elfmannen zijn bezig Sokrates te ontboeien en kondigen hem aan,
hoe hij dezen dag zal moeten sterven. Niet langen tijd nu wachtte hij
60 van te komen en verzocht ons binnen te gaan. Wij traden
derhalve binnen en vonden Sokrates pas ontboeid en Xanthippe—gij
kent haar immers?—naast hem gezeten, met zijn zoontje bij zich.
Toen nu Xanthippe ons zag, jammerde zij luide en zeide dergelijke
dingen als vrouwen gewoon zijn: o Sokrates, dit is nu de laatste
maal, dat gij en uw vrienden elkander zult toespreken! En Sokrates
zag naar Kritoon en zeî: o Kritoon, laat iemand haar naar huis
brengen. En haar leidden enkele van Kritoons mannen weg, terwijl
B zij luid weende en rouwmisbaar maakte. Onderhand ging
Sokrates op zijn bed overeind zitten en trok zijn been op en wreef
het met de hand, en zeide al wrijvende: Wat een vreemd ding, o
mannen, lijkt dat te zijn wat de menschen ’t aangename noemen!
Hoe wonderlijk verhoudt het zich van-nature tegenover zijn
algemeen erkend tegengestelde, het smartelijke, dat beide tegelijk
niet bij den mensch willen wezen, maar dat als men het éene
najaagt en krijgt, men zoo goed als gedwongen wordt ook het
tweede te krijgen, alsof zij hoewel zij twee blijven, met éen toppunt
C aan elkander zijn vastgemaakt. En mij komt het voor, zeide hij, dat
Aisopos, als hij op hen bedacht ware geweest, er een fabel van zoû
samengesteld hebben, dat de godheid met de bedoeling om hen in
hun vijandschap te verzoenen, toen hij dat niet kon, hunne
toppunten aan-éen heeft geknoopt, en dat daarom bij wien het éene
gekomen is, later ook het tweede volgt. Zooals het nu ook bij mij
zelven lijkt: nadat tengevolge van de boei in mijn been het pijnlijke
was, blijkt nu daar-achteraan-vast te komen het aangename.
Kebes nam nu het woord en zeide: Bij Zeus, o Sokrates, dat is
D goed van u, dat gij er mij aan herinnert! Want omtrent de verzen
toch, die gij gemaakt hebt door de fabelen van Aisopos op maat te
brengen en het proëem te dichten op Apolloon, hebben mij sommige
anderen reeds gevraagd, doch een paar dagen geleden ook
Euenos, met welke bedoeling gij, nadat gij hier gekomen waart, hen
vervaardigd hebt, terwijl gij vroeger nog nooit iets gedicht hebt. Als
gij er derhalve belang in stelt, dat ik Euenos antwoord kan geven,
wanneer hij het weêr vraagt—want ik weet wel dat hij het vragen zal
—, zeg mij wat ik moet zeggen.—Zeg hem dan de waarheid, o
Kebes, dat ik die niet gemaakt heb met den wil om een kunst-
E mededinger te zijn van hem en zijne gedichten. Want ik wist dat
dit niet gemakkelijk zoû zijn. Maar om zekerheid te hebben omtrent
de bedoeling van bepaalde droomen en een onbezwaard geweten te
hebben, indien misschien dit de moezische kunst was, welke die
droomen mij herhaaldelijk opdroegen te vervaardigen. Zij waren
namelijk van den volgenden aard. Vele malen bezocht mij dezelfde
droom in mijn voorbije leven, nu onder dit, dan onder dat gezicht
verschijnende, maar met dezelfde boodschap: o Sokrates,
vervaardig moezische kunst en houd u daarmeê bezig. En ik
61 veronderstelde in den voorafgaanden tijd, dat de droom mij
aanspoorde en bijval toeriep om dat te doen waarmeê ik bezig was,
evenals toeschouwers in de renbaan de loopers aanmoedigen,—dat
ook zoo de droom mij toejuichend aanspoorde te doen wat ik al
deed, moezische kunst maken, daar philosofie, als ik meende, de
grootste moezische kunst was, en ik mij daarmede bezig hield. Maar
nu, nadat het proces afgeloopen was, en tegelijk het feest van den
god verhinderde dat ik stierf, meende ik dat, indien de droom mij
weder herhaaldelijk mocht opdragen die moezische kunst, nu-dan
naar de gemeene volksopvatting, te vervaardigen, ik den droom niet
ongehoorzaam mocht wezen, maar behoorde te dichten. Want ik
B dacht, dat het veiliger was niet heen te gaan, vóor ik mij gereinigd
had van de verplichting om gedichten te maken, en zoo den droom
gehoorzaam te zijn. Zoo heb ik dan in de eerste plaats een lied
gemaakt op den god wiens offerfeest het was, en na den god heb ik,
bedenkende dat een dichter, zal hij werkelijk dichter zijn, verzinselen
en geen feiten behoort te dichten, omdat ik zelf geen fabelverzinner
was, de fabelen die ik bij de hand had en uit mijn hoofd kende, de
fabelen van Aisopos, op maat gebracht, de eerste waar ik op kwam.
Zeg dit derhalve, o Kebes, aan Euenos, en dat het hem welga, en
C dat hij, als hij verstandig is, mij zoo spoedig mogelijk volge. Want
ik zal, naar het schijnt, vandaag heengaan. De Atheners toch
gebieden het.—En Simmias zeide: Wat is dat voor eene aanmaning,
o Sokrates, aan Euenos?... Want reeds vele malen heb ik den man
ontmoet. Uit wat ik nu waargenomen heb, is het vrijwel zeker, dat hij
uit-eigen-wil geenerwijs uw raad zal opvolgen.—Wat zegt gij daar? Is
Euenos geen wijsgeer?—Mij dunkt van-wel, zeide Simmias.—Dan
zal hij dat gaarne doen, zoowel Euenos als ieder die daaraan
waardiglijk deel heeft. Toch zal hij niet waarschijnlijk geweld aan
zichzelf plegen; want dat is niet geoorloofd, zegt men.—Onder ’t
D zeggen van deze woorden liet hij zijn beenen op den grond neêr,
en van dat oogenblik voerde hij zoo gezeten het gesprek.—Kebes
nu vroeg hem: Hoe bedoelt gij dit, Sokrates, dat het niet geoorloofd
is geweld aan zichzelf te plegen, maar dat toch de wijsgeer met den
stervende graag zoû meêgaan?—Wat, Kebes, hebt gij en Simmias
die bij Philolaos in de leer zijt geweest, hieromtrent niet gehoord?—
Niets zekers tenminste, o Sokrates.—Ook ik-voor-mij praat er maar
van-hooren-zeggen over. Doch wat ik bij-geval gehoord heb, heb ik
geen enkele reden te verzwijgen. Immers, past het wel, vooral nu ik
E op het punt sta daarheen af te reizen, eens verstandelijk na te
gaan en in woorden uit te beelden, wat-voor meening wij hebben
omtrent de reis daarheen. En ook—wat zoû men anders doen in den
tijd tot den ondergang der zon?
—Op welken grond dan toch wel beweert men, dat het niet
geoorloofd is zichzelf te dooden, o Sokrates? Want ik heb (waarnaar
gij mij zoo-even vraagdet) ook van Philolaos, toen die bij ons
verkeerde, en ook al van verscheidene anderen gehoord, dat men
dit niet mag doen. Maar iets beredeneerd-nauwkeurigs heb ik nog
62 nooit van iemand erover gehoord.—Houd maar goeden moed,
zeide hij, want misschien zult gij het nog wel hooren.... Het zal u
evenwel denkelijk wonderlijk schijnen, dat dit alleen onder alle
dingen eenvoudig vaststaat en nooit voor den mensch van
omstandigheden afhangt zooals de andere dingen wel: het zal u
misschien wonderlijk voorkomen, zeg ik, dat, hoewel het
verscheidenen op verscheidene tijden beter is dood te zijn dan te
leven, het dien menschen wien het beter is dood te zijn, niet oorbaar
is zichzelf wel te doen, maar zij een anderen weldoener moeten
afwachten.—En Kebes glimlachte even en: Dat weet Deus, zeide hij
B in zijn eigen tongval.—Het zoû ook om-zoo-te hooren, zeide
Sokrates, onverklaarbaar kunnen klinken, maar niettemin heeft het
wellicht eenigen grond. De verklaring nu die ons hieromtrent in de
oude mysteriën gegeven wordt, dat wij menschen in een soort van
gevangenschap zijn, en dat men zich uit deze niet mag bevrijden en
niet wegloopen, lijkt mij een groote-bewering te zijn en niet
gemakkelijk om te doorzien. Doch in-elk-geval schijnt mij het
volgende terecht beweerd te worden, dat het de goden zijn, die voor
ons zorg dragen, en dat de menschen voor de goden een hunner
C bezittingen zijn. Of denkt gij er niet zoo over?—Ja zeker, zeide
Kebes.—Zoudt nu ook gij niet, indien een uwer slaven zichzelf
doodde zonder aanwijzing van u dat gij dit wildet, boos op hem zijn
en hem straffen als gij kondt?—Voorzeker.—Misschien is het dan op
dezelfde wijze niet ongegrond, dat men zelf zich niet eerder mag in
den dood begeven, vóor de godheid een-of-andere noodzakelijkheid
op ons afzendt, zooals nu in ons geval.—Wel, dit althans komt mij
waarschijnlijk voor. Wat gij evenwel daar-straks zeidet, dat de
D wijsgeeren reede zouden willen sterven, dat lijkt vreemd,
Sokrates, indien wat wij zoo-juist beweerden steek houdt, dat het
namelijk de godheid is die voor ons zorgdraagt, en dat wij zijne
bezittingen zijn. Want dat het den verstandigsten geen leed zoû
doen heen te gaan uit deze dienstbaarheid waarin over hen gesteld
zijn die de beste toezieners zijn van alle bestaande, de goden, heeft
geen zin. Want zoo-een meent allicht niet, dat hij, vrij geworden, zelf
beter voor zich zorgen zal. Maar een ònverstandig mensch
waarschijnlijk zal dit denken, dat hij weg moet loopen van zijn heer,
E en zal niet berekenen dat men van zijn góeden heer tenminste
niet mag wegloopen, maar boven-al bij hem behoort te blijven,
omdat het onzinnig zoû zijn weg te loopen. Doch een verstandig
man zal allicht begeeren voor altijd bij zijnen betere te zijn. En toch,
zoo schijnt het tegenovergestelde waar te zijn van wat daarnet
beweerd werd: de verstandigen behooren bedroefd te zijn bij hun
sterven, de onverstandigen blij.—Op het hooren hiervan had
63 Sokrates, kwam mij voor, genoegen in Kebes’ gevatheid. Hij
wierp ons een blik toe en zeide: Altijd toch spoort Kebes een-of-
andere tegenwerping op, en wil zich nooit erg gauw laten overtuigen
van wat men ook beweert.—En Simmias sprak: Maar nu toch, o
Sokrates, vind ik ook zelf, dat Kebes iets van belang beweert. Want
met welke bedoeling zouden naar-waarheid wijze mannen heeren
die hun beteren zijn, willen ontloopen en zich lichtvaardig van hen
ontdoen? Ook komt het mij voor, dat Kebes in zijn woorden op ù
B doelt, dat gij het zoo licht opneemt, zoowel om ons te verlaten als
goede bestuurderen, zooals gij zelf toegeeft, de goden.—Gij vraagt
uw recht, zeide hij. Want uwe bedoeling is, meen ik, dat ik mij
daaromtrent behoor te verdedigen als voor een gerecht.—Ja juist,
zeide Simmias.
—Welaan dan, zeide hij, laat ik trachten met meer overtuigenden
uitslag mij tegenover u te verdedigen dan tegenover mijne eerste
rechters. Indien ik namelijk, o Simmias en Kebes, niet meende te
zullen komen vooreerst tot andere goede en wijze goden, en verder
tot gestorven menschen, betere dan die hier zijn, zoude het zondig
van mij wezen niet bedroefd te zijn om mijn dood. Maar nu, weet
C wel, dat ik hoop te zullen komen tot goede menschen,—en dit wil
ik nog niet zoo zeker beweren,.... dat ik evenwel tot goden zal
komen, die zeer goede heeren voor mij zullen zijn, weet wel dat ik,
zoo eenig ander ding van dien aard, dit zoû durven verzekeren.
Zoodat ik daarom geen reden heb bedroefd te zijn, maar welgemoed
ben, dat er voor de menschen na hun dood nog iets bestaat, en
zooals ook vanouds overgeleverd wordt, iets dat veel beter is voor
de goeden dan voor de slechten.—Hoe dan, Sokrates, zeide
D Simmias, zijt gij van-plan heen te gaan en deze overtuiging voor
uzelf te houden, of zult gij ons haar mededeelen? Want het komt mij
voor, dat dit een goed is, waar ook wij recht op hebben, en tegelijk
zal het u voor verdediging gelden, wanneer gij ons overtuigt van wat
gij zegt.—Ik zal het beproeven, zeide hij. Maar laten wij eerst zien
wat Kritoon hier wil. Het komt mij voor, dat hij al geruimen tijd iets te
zeggen had.—Niets anders, o Sokrates, zeide Kritoon, dan dat de
man die u het gif zal toedienen, mij voortdurend zegt, dat ik u moet
beduiden zoo weinig mogelijk te praten. Want, zegt hij, bij het praten
maken de menschen zich warm meer dan goed is, en zoo-iets mag
E heelemaal niet gebeuren bij het drinken van den giftbeker, en als
men niet naar hem luistert, zijn menschen die zoo-iets doen,
somtijds gedwongen twee-, ja driemaal te drinken.—En Sokrates
zeide: Laat hem praten. Als hij maar zijn eigen werk in orde brengt,
alsof hij dan tweemaal, of als het noodig mocht zijn, zelfs driemaal
zal moeten toedienen.—Ik wist uw antwoord al vooruit, zeî Kritoon.
Maar de man is lastig.—Laat hem, zeide hij.
Maar tegenover u dan, mijne rechters, wil ik nu mijne
verantwoording afleggen, dat ik op goede gronden meen, dat een
man die zijn leven inderdaad in wijsbegeerte heeft doorgebracht,
64 goedsmoeds is op het oogenblik van zijn dood en goederhoop,
dat hij daarginds de grootste goede dingen zal behalen, wanneer hij
gestorven is. Hoe dit zoo komt te zijn, o Simmias en Kebes, zal ik
trachten u duidelijk te maken. Het heeft er namelijk veel van, dat van
allen zoovelen zich ordelijk toeleggen op wijsbegeerte, de overigen
niet bemerken, dat zij-voor-zich niets anders bestreven dan sterven
en dood-zijn. Indien dit nu de waarheid is, zoû het natuurlijk
ongerijmd zijn van hen, in hun geheele leven niets anders dan dit te
bestreven en dan, wanneer juist datgene komt, waarvan zij al-lang
hun dagelijksch streven maakten, mistroostig te zijn.—En Simmias
B zeide lachend: Bij Zeus, o Sokrates, ik ben op het oogenblik niet
erg tot lachen gestemd, maar gij doet mij lachen. Want ik meen, dat
de gemeene luiden die dit hoorden, zouden vinden, dat juist dit erg
van pas gezegd was aan het adres der wijsgeeren, en het zouden
beamen, de menschen bij ons vooral, dat inderdaad de wijsgeeren
naar den dood hunkeren, en verdienen zoo te varen, en dat zij dit
heel goed van hen merken.—Dan zouden zij de waarheid zeggen, o
Simmias, behalve de bewering, dat zij het zouden merken. Want het
ontgaat hun, hoe naar den dood hunkeren en hoe en welken dood
C verdienen de waarachtige philosofen. Doch laten wij dien anderen
vaarwel zeggen en praten alleen lettend op onszelven.
Meenen wij, dat de dood iets bepaalds is?—Zeker, gaf Simmias
daarop ten antwoord.—Wel iets anders dan de scheiding der ziel
van het lichaam? En dat dood-zijn dit is, dat éenszijds het lichaam
van de ziel afgescheiden iets op-zich-zelfs komt te zijn, en
anderszijds de ziel van het lichaam gescheiden op-zich-zelf is?
Bestaat er soms kans, dat de dood iets anders zoû zijn dan dit?—
Neen, dit alleen, zeide hij.—Ga dan even na, mijn waarde, of ook gij
over het volgende denkt als ik. Want hiervan uitgaande zullen wij tot
D beter inzicht komen in de dingen waarover wij redeneeren.... Lijkt
het u eens wijsgeers bezigheid te zijn ernst te maken met zulke
zoogenaamde genietingen als die van eten en drinken?—Allerminst,
o Sokrates, zeî Simmias.—En verder de lusten-in-liefde?—
Geenerwijs.—En verder,.... lijkt u de zoodanige de overige
dienstplegingen betreffende het lichaam van-waarde te houden, als
de verwervingen van keur-onderscheiden opperkleederen en
schoeiselen en de overige uiterlijk-vermooiingen betreffende het
E lichaam?.... denkt gij, dat hij ze van-waarde zal achten of
waardelóos voor-zoo-ver er geen groote noodzakelijkheid bestaat er
deel aan te hebben?—Ik denk, dat hij ze waardeloos acht, tenminste
de waarachtige wijsgeer.—Dunkt u dan niet over-’t-geheel de
leefbezigheid van den zoodanige niet te gaan om het lichaam, maar
zooveel als kan, zich daarvan op een afstand te houden en naar de
zijde der ziel gericht te zijn?—Ja.—Valt het derhalve soms (in de
65 eerste plaats in dergelijke zaken) van den wijsgeer in ’t licht, dat
hij zooveel mogelijk de ziel tracht los te maken van hare
gemeenschap met het lichaam op een van de overige menschen
verschillende wijze?—Het schijnt zoo.—En daarom natuurlijk lijkt
het, o Simmias, aan de groote menigte der menschen, dat hij wien
niets van dergelijke dingen genot geeft en die geen deel aan hen
heeft, niet waard is te leven, maar dat heel dicht aan dood-zijn reikt
de man die zich niet kommert om de lusten die door het lichaam
heen ons bereiken.—Gij zegt de volle waarheid.— —Maar hoe staat
het dan met de verwerving zelve van inzicht? Is het lichaam een
B voetstoot of niet, wanneer men het bij het zoeken daarnaar als
speurgenoot met zich medeneemt? Ik bedoel bij-voorbeeld het
volgende: Heeft gezicht en gehoor voor de menschen wel eenige
waarachtigheid? Of zijn maar praatjes-voor-den-vaak, ook van de
dichters, diergelijke beweringen die zij ons aldoor houden, dat wij
met nauwkeurigheid noch iets hooren noch zien? Toch, als van de
lichamelijke waarnemingen deze niet nauwkeurig en niet zeker zijn,
zijn de andere het niet-licht; want alle zijn zij wel gebrekkiger. Of
vindt gij van-niet?—Neen, bepaald van-wel.—Wanneer derhalve
beroert de ziel de waarheid? Wanneer zij immers in gezelschap van
het lichaam iets tracht te onderzoeken, dan wordt zij, dat is duidelijk,
C daardoor misleid.—Gij spreekt de waarheid.—Wordt derhalve
niet, zoo ergens anders, in het denken iets van de werkelijkheid haar
eigen?—Ja.—En denkt zij misschien dan schóonst, wanneer niets
van deze dingen haar hindert, noch gehoor, noch gezicht, noch pijn,
noch eenige lust, maar wanneer zij zooveel mogelijk op-zich-zelf
komt te zijn en het lichaam aan zijn lot overlaat en, zooveel zij kan,
daar geen deel aan neemt en er niet meê in aanraking is, en dan
naar de waarheid reikt?—Dat is zoo.—Schat derhalve ook daarbij de
ziel van den wijsgeer het lichaam niet ten zeerste waardeloos en
D vlucht er vandaan en zoekt op-zich-zelf alleen te zijn?—Dat is
helder.——En wat dunkt u verder van het volgende: Beweren wij dat
het begrip rechtvaardig iets bestaands is of niets?—Voorzeker bij
Zeus.—En het begrip schoon en het begrip goed?—Hoe niet?—Hebt
gij dan wel ooit eenig van diergelijke dingen met uw oogen gezien?
—Geenszins, zeide hij.—Maar hebt gij met eenige andere der door
het lichaam heengaande waarnemingen hen beroerd? Ik bedoel alle
E dingen als grootte, gezondheid, kracht, in-éen-woord het wezen
van de gezamenlijke overige dingen, wat elk afzonderlijk eigenlijk is.
Wordt door het lichaam heen hun waarachtigste wezen
aanschouwd? Of is het er zóo meê: wie van ons zich toelegt om het
wezen van elk ding afzonderlijk waar hij onderzoek naar doet, meest
en nauwgezetst met de gedàchte te doorgronden, zal díe wel het
naast komen tot de bewustwording van elk afzonderlijk?—Volslagen
zoo.—Zal derhalve die man dat het zuiverst doen, die zooveel
mogelijk met de gedachte zelve op elk ding afgaat zonder het
66 gezicht bij het doordenken erbij te nemen, noch eenig ander
zintuig bij de overweging erbij te halen, maar die met de
zonnezuivere gedachte op-zich-zelf ieder ding zonnezuiver op-zich-
zelf tracht na te jagen van de bestaande dingen, na zich zooveel
mogelijk ontdaan te hebben van oogen en ooren en om-zoo-te-

You might also like