Download Full Version ebookmass - Visit ebookmass.

com

Visual Data Insights Using SAS ODS Graphics: A

Guide to Communication-Effective Data
Visualization 1st Edition Leroy Bessler

https://ebookmass.com/product/visual-data-insights-using-
sas-ods-graphics-a-guide-to-communication-effective-data-
visualization-1st-edition-leroy-bessler/

OR CLICK HERE

DOWLOAD NOW

Discover More Ebook - Explore Now at ebookmass.com

 Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...

Visual Data Insights Using SAS ODS Graphics: A Guide to

Communication-Effective Data Visualization 1st Edition
Leroy Bessler
https://ebookmass.com/product/visual-data-insights-using-sas-ods-
graphics-a-guide-to-communication-effective-data-visualization-1st-
edition-leroy-bessler/
ebookmass.com

Data Science With Rust: A Comprehensive Guide - Data

Analysis, Machine Learning, Data Visualization & More Van
Der Post
https://ebookmass.com/product/data-science-with-rust-a-comprehensive-
guide-data-analysis-machine-learning-data-visualization-more-van-der-
post/
ebookmass.com

Text as Data: Computational Methods of Understanding

Written Expression Using SAS (Wiley and SAS Business
Series) 1st Edition Deville
https://ebookmass.com/product/text-as-data-computational-methods-of-
understanding-written-expression-using-sas-wiley-and-sas-business-
series-1st-edition-deville/
ebookmass.com

What Matters in Survival: Personal Identity and other

Possibilities 1st Edition Douglas Ehring

https://ebookmass.com/product/what-matters-in-survival-personal-
identity-and-other-possibilities-1st-edition-douglas-ehring/

ebookmass.com
Extended finite element and meshfree methods 1st Edition
Timon Rabczuk

https://ebookmass.com/product/extended-finite-element-and-meshfree-
methods-1st-edition-timon-rabczuk/

ebookmass.com

Nutrition For Healthy Living 4th Edition, (Ebook PDF)

https://ebookmass.com/product/nutrition-for-healthy-living-4th-
edition-ebook-pdf/

ebookmass.com

Introduction to the Practice of Statistics 9th Edition,

(Ebook PDF)

https://ebookmass.com/product/introduction-to-the-practice-of-
statistics-9th-edition-ebook-pdf/

ebookmass.com

The Unexpected in Oral History Ricardo Santhiago

https://ebookmass.com/product/the-unexpected-in-oral-history-ricardo-
santhiago/

ebookmass.com

Industrial/Organizational Psychology: An Applied Approach,

9e 9th Edition Michael G. Aamodt

https://ebookmass.com/product/industrial-organizational-psychology-an-
applied-approach-9e-9th-edition-michael-g-aamodt/

ebookmass.com
The Music Professor Online Judith Bowman

https://ebookmass.com/product/the-music-professor-online-judith-
bowman-2/

ebookmass.com
Visual Data
Insights Using
SAS ODS
Graphics
A Guide to Communication-
Effective Data Visualization
―
LeRoy Bessler
 VISUAL DATA INSIGHTS
USING SAS ODS GRAPHICS

A GUIDE TO COMMUNICATION-EFFECTIVE
DATA VISUALIZATION

LeRoy Bessler
Visual Data Insights Using SAS ODS Graphics: A Guide to Communication-
Effective Data Visualization

LeRoy Bessler
Mequon, WI, USA

ISBN-13 (pbk): 978-1-4842-8608-1 ISBN-13 (electronic): 978-1-4842-8609-8

https://doi.org/10.1007/978-1-4842-8609-8
Copyright © 2023 by LeRoy Bessler
This work is subject to copyright. All rights are reserved by the Publisher, whether the
whole or part of the material is concerned, specifically the rights of translation, reprinting,
reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other
physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter
developed.
Trademarked names, logos, and images may appear in this book. Rather than use a
trademark symbol with every occurrence of a trademarked name, logo, or image we use
the names, logos, and images only in an editorial fashion and to the benefit of the trademark
owner, with no intention of infringement of the trademark.
The use in this publication of trade names, trademarks, service marks, and similar terms,
even if they are not identified as such, is not to be taken as an expression of opinion as to
whether or not they are subject to proprietary rights.
While the advice and information in this book are believed to be true and accurate at the
date of publication, neither the authors nor the editors nor the publisher can accept any
legal responsibility for any errors or omissions that may be made. The publisher makes no
warranty, express or implied, with respect to the material contained herein.
Managing Director, Apress Media LLC: Welmoed Spahr
Acquisitions Editor: Susan McDermott
Development Editor: Laura Berendson
Coordinating Editor: Jessica Vakili
Distributed to the book trade worldwide by Springer Science+Business Media New York,
233 Spring Street, 6th Floor, New York, NY 10013. Phone 1-800-SPRINGER, fax (201)
348-4505, e-mail [email protected], or visit www.springeronline.com. Apress
Media, LLC is a California LLC and the sole member (owner) is Springer Science + Business
Media Finance Inc (SSBM Finance Inc). SSBM Finance Inc is a Delaware corporation.
For information on translations, please e-mail [email protected]; for
reprint, paperback, or audio rights, please e-mail [email protected].
Apress titles may be purchased in bulk for academic, corporate, or promotional use. eBook
versions and licenses are also available for most titles. For more information, reference our
Print and eBook Bulk Sales web page at http://www.apress.com/bulk-sales.
Any source code or other supplementary material referenced by the author in this book is
available to readers on the Github repository: https://github.com/Apress/Visual-Data-
Insights-Using-SAS-ODS-Graphics. For more detailed information, please visit http://
www.apress.com/source-code.
Printed on acid-free paper
To creators of data visualization, the intended readers,
and
to William S. Playfair, Philippe Buache, and
Guillaume de L’Isle
whose graphic innovations in the late 1700’s in
England and in France
remain the most widely used and frequently used ways
to provide and find visual data insights
Let your computer draw a picture
to let viewers see the data
with image for an easy, immediate impression
of what’s larger, what’s smaller
of what the trend is
of what the relationship is
of what the distribution is
including geographically
and with precise numbers for correct, reliable understanding.
Let it paint a picture
that shows the viewer what’s important.
Contents
About the Author ix
About the Technical Reviewer xi
Acknowledgments xiii
Introduction xv

Part I: Design Principles 1

Chapter 1: Principles of Communication-Effective Graphic Design 3
Chapter 2: Principles of Communication-Effective Use of Color 31

Part II: Widely Applicable Examples You Can Use 53

Chapter 3: Introduction to SAS ODS Graphics 55
Chapter 4:  ar Charts, Butterfly Charts, Waterfall Charts,
B
Dot Plots, Needle Plots, Area Bar Charts, Text Graphs,
and Line Charts: Charts for Categorical Data 85
Chapter 5: Pie Charts and Donut Charts 187
Chapter 6: Heat Maps 221
Chapter 7: Bubble Plots 263
Chapter 8: Time Series Plots and Trend Lines 283

Part III: Other Features 355

Chapter 9: Graphic Composites with PROC SGPANEL 357
Chapter 10: S
 catter Plots in Composites Using
PROC SGSCATTER 419
Chapter 11: Fits and Confidence Plots 447
Chapter 12: D
 istributions, Histograms, Box Plots, and
Alternative Tools 465
Chapter 13: C
 reating Composites of Graphs, Tables, and Text
with ODS LAYOUT 503
viii Contents

Chapter 14: D
 elivering Precise Numbers and Alternative
Views for Graphs Using SAS ODS HTML5 525
Chapter 15: D
 elivering Precise Numbers When Using
PROC SGMAP 573
Appendix A 581
For Further Information on SAS ODS Graphics 607

Index 611
About the Author
LeRoy Bessler is a data analyst, a SAS soft-
ware expert, a data visualization aficionado
since 1981, and an advocate for and demonstra-
tor of his graphic design principles since 1991.
He earned a Ph.D. in physics with a minor in
mathematics from the University of Wisconsin-
Milwaukee. He served as Assistant Professor of
Mathematics at Milwaukee School of Engineering
and later was appointed Senior Fellow in
Theoretical Physics at Queen Mary College,
University of London. When finding the theory
of submicroscopic elementary particles becom-
ing, to his taste, increasingly unphysical, LeRoy
returned to the United States to devote his
interest and energy to the macroscopic needs of American business using
computers and networks, working for employers and clients in finance, health
insurance, property and casualty insurance, manufacturing, energy, and retail.
After a variety of roles, responsibilities, and accomplishments in information
technology, he decided to concentrate on using SAS software as his IT tool.
LeRoy has supported SAS servers, SAS software, SAS data, and the users of
those facilities, as well as working as a data analyst and SAS programmer. His
distinguishing expertise with SAS software has been communication-effective
data visualization, and software-intelligent application development for reli-
ability, reusability, maintainability, extendability, and flexibility—to deliver
Visual Data Insights™ and Strong Smart Systems™.
Still a professor at heart, LeRoy has shared his ideas, knowledge, and experi-
ence at conferences for SAS users throughout the United States and in
Europe. With the help of volunteers, he has enjoyed creating SAS user mutual
education opportunities. LeRoy is a How To contributor to VIEWS News, the
online quarterly for SAS users.
He has served as an elected and appointed local government official and on
the boards of social services, cultural, and civic nonprofit organizations.
About the Technical
Reviewer
Philip R. Holland has over 30 years of experience of working with SAS
software. In 1992, he formed his own consultancy company, Holland Numerics
Ltd. Since then, he has provided SAS technical consultancy and training in the
financial, retail, and pharmaceutical sectors in the UK, Belgium, Holland,
Germany, and the United States. Since 1995, he has presented papers on a
wide range of SAS-related topics at SAS user conferences and seminars in the
UK, Europe, and the United States and has published four SAS-related books
and eight SAS-related courses.
While writing his thesis for a doctorate in chemistry in the 1980s, he gave an
early draft to a colleague to read, and they found 60 errors in spelling and
grammar in the first chapter! This made him realize that no one can proofread
their own work, because you read what you think is there, rather than what
you have actually written. Since then, he has helped other book and SAS pro-
gram writers avoid his own proofreading embarrassment.
Acknowledgments
I am immensely grateful to executive editor Susan McDermott for engaging
me to write this book, to Apress editors Jessica Vakili, Rita Fernando, and
Laura C. Berendson, to Apress production coordinator Krishnan Sathyamurthy,
to Apress project manager Angel Michael Dhanaraj, and to Philip R. Holland
who did the technical review. It was Phil who got me connected to Apress.
Without these people, and the rest of the Apress team, there would have
been no book.
I am indebted to Marcia Surratt, Lelia McConnell, Martin Mincey, Kathryn
McLawhorn, Amber Elam, Cyrus Bradford, and Liz Edwards at SAS Technical
Support who handled my problems and questions during this project.
Alan Paller, an analyst in the early days of the computer graphics industry,
encouraged me to become the advocate for graphics at Miller Brewing
Company. There, at my suggestion, Thomas S. Cain added support of all three
graphics software products to my workplace responsibilities. Atis Purins com-
missioned me do a makeover of the monthly report to Miller management on
usage, capacity, and performance for all of the information technology facili-
ties. That project not only engaged my graphic design principles, but also
inspired me to implement what I call Software-Intelligent Application
Development. SIAD assures that the code can auto-adapt to changes in the
run-time environment to maintain the design objectives. Chris Potter, the
1990 SAS Users Group International Conference Graphics Section Chair,
liked what he saw in one of my presentations and encouraged me to promote
my design principles at future conferences. That is how my journey of devel-
oping and sharing my ideas, knowledge, and experience with SAS software for
data visualization began.
Over decades, numerous conference organizers kindly provided me so many
opportunities to write and speak about SAS topics, including data visualiza-
tion. That engendered my development of an ever-growing list of design
guidelines for graphics and color use, and a portfolio of examples to illustrate
application of those guidelines.
I thank David V. Evans of J.C. Penney who hired a repatriated computer know-
nothing as an information technology trainee, starting me on a new career.
That career ultimately got me to this book.
Carol Bessler gets me through life. Thank you.
Introduction
The visual data insight provided by graphics is essential to understand data.
Statistics alone clearly are not sufficient, as demonstrated with Anscombe’s
quartet of data sets, for which regressions are plotted in Figure 1. All four
very different images have statistics that are nearly identical. See the four
tables in Table 1 and further remarks there.
This book is an experience-based, practical handbook for applying communi-
cation-effective design principles (48 for graphs and 23 for color), as demon-
strated in 327 examples.
The data visualization tool used is SAS ODS Graphics, but the principles are
software independent, relevant for any tool.
Graphs can make it unnecessarily difficult to understand the data, can confuse
the viewer, or can mislead the viewer. If a graph needs an explanation, it has
failed to communicate.
Misuse of color (or colour in some countries) likewise has adverse visual con-
sequences. I’ve seen—no, I struggled to see—yellow text or yellow markers
on a white background and black text on dark blue. The only options worse
could be white on white and black on black. Another common problem is the
use of continuous color gradients for color coding. Determining exactly which
colored areas are the same color is impossible, and matching an area color
with its corresponding color along a continuous color gradient legend is like-
wise. Those are not the only or the worst unwise uses of color.
This book explains and demonstrates how to get the best out of ODS
Graphics, relying on numerous guidelines. Three deserve special emphasis.
First, though images are for quick, easy inference, the associated precise num-
bers are needed for correct, reliable inference. The book shows all of the
ways to make them part of the image. Moving the eye from a graphic element
(bar end or plot point) to axis tick mark values and estimating is not a solution.
Let your data talk. Show and Tell. Data can show its behavior with the visual—
which category is bigger/smaller or where it’s going over time—and it can tell
the viewer its values with annotation, an axis table, or web-enabled
mouseover text.
xvi Introduction

Second, avoid information overload. Show the Viewer What’s Important. Use
ranking and subsetting. Three ways of subsetting are shown in Figure 4-7.
Another way is via sparse presentation, as in Figures 1-18 and 1-15. Ranking
is a huge help, is used in numerous examples, and requires little code.
Third, both readability of the text and usability of any color coding are usually
taken for granted, but must be assured. See the recommendations.
The principles in this book are application independent, and the examples can
be adapted to any data for any industry, enterprise, or organization.
Code for all of the examples can be downloaded at https://github.
com/Apress/Visual-Data-Insights-Using-SAS-ODS-Graphics. Some
examples use %INCLUDE statements to retrieve macros or
included code blocks. A zip file of includables must be downloaded.
If your site does not have SAS/ETS, a zip file of needed ETS data
sets can be downloaded. The overwhelming majority of examples
use SASHELP data sets that all sites have.
You can re-create any example, and experiment with options, or apply the
example code to your own data, adapting the code however desired.
The book’s scope is anything that ODS Graphics can do. Though the exam-
ples cannot show you everything, they do span that range.
As I traveled through the range, applying recommended principles to
representative examples, I also built alternatives that are unexpected, but
improved. The improvements are too many to enumerate here. An old and
frequently repeated (not by me) criticism of pie charts is refuted—three ways.
And the simplest possible pie chart is demonstrated as a very powerful visual.
You will see innovations. For statistics, there are new ways to create distribu-
tions, box plots, and histograms. For categorical data, there is the Tree Chart,
Flag Chart, CrossRoads SignPost Chart, and bar chart with no bars. For time
series data, there is the Sparse (not spark) Line—alone, in tables, in panels,
and web enabled—and also a trend line with no line.
Three different ODS Graphics features to create composites of graphs and/or
tables are covered. One way is web enabled, with the added benefit of pop-up
mouseover text, which is indispensable in cases where permanent annotation
of plot points is infeasible.
This book is the culmination and, in effect, an illustrated biography of my
working, learning, and discovery with SAS graphics software since 1981. My
conclusions about best design with graphics and color gradually grew longer
and longer. I have applied them with the benefit of experience, now using SAS
ODS Graphics to create images, illuminations, and insights for data, which are
correct, clear, concise but complete, convincing, compelling, and, when
needed or otherwise appropriate, colorful. Join me in the quest.
LeRoy Bessler
 Introduction xvii

Figure 1. Demonstration of the need for and value of data visualization

I learned about the four data sets from an article by Philip R. Holland in Issue
54 (2nd/3rd Quarter 2011) of VIEWS News. For more information about
Francis J. Anscombe’s data sets, see https://en.wikipedia.org/wiki/
Anscombe's_quartet. For the example of The Datasaurus Dozen, see www.
autodesk.com/research/publications/same-stats-different-graphs.
xviii Introduction

Table 1. SAS REG procedure statistics for the Anscombe quartet data

Model: MODEL1 Dependent Variable: y

Number of Observations Read 11, Number of Observations Used 11
DataSet=I ----------------------------------------------------------------

Analysis of Variance
Source DF Sum of Squares Mean Square F Value Pr > F
Model 1 27.51000 27.51000 17.99 0.0022
Error 9 13.76269 1.52919
Corrected Total 10 41.27269
Root MSE 1.23660 R-Square 0.6665
Dependent Mean 7.50091 Adj R-Sq 0.6295
Coeff Var 16.48605

Parameter Estimates
Parameter Standard
Variable DF Estimate Error t Value Pr > |t|
Intercept 1 3.00009 1.12475 2.67 0.0257
x 1 0.50009 0.11791 4.24 0.0022

DataSet=II --------------------------------------------------
Analysis of Variance
Source DF Sum of Squares Mean Square F Value Pr > F
Model 1 27.50000 27.50000 17.97 0.0022
Error 9 13.77629 1.53070
Corrected Total 10 41.27629
Root MSE 1.23721 R-Square 0.6662
Dependent Mean 7.50091 Adj R-Sq 0.6292
Coeff Var 16.49419

Parameter Estimates
Parameter Standard
Variable DF Estimate Error t Value Pr > |t|
Intercept 1 3.00091 1.12530 2.67 0.0258
x 1 0.50000 0.11796 4.24 0.0022

(continued)
 Introduction xix

Table 1. (continued)
DataSet=III ---------------------------------------------------------------
Analysis of Variance
Source DF Sum of Squares Mean Square F Value Pr > F
Mode 1 27.47001 27.47001 17.97 0.0022
Error 9 13.75619 1.52847
Corrected Total 10 41.22620
Root MSE 1.23631 R-Square 0.6663
Dependent Mean 7.50000 Adj R-Sq 0.6292
Coeff Var 16.48415

Parameter Estimates
Parameter Standard
Variable DF Estimate Error t Value Pr > |t|
Intercept 1 3.00245 1.12448 2.67 0.0256
x 1 0.49973 0.11788 4.24 0.0022

DataSet=IV ----------------------------------------------------------
Analysis of Variance
Source DF Sum of Squares Mean Square F Value Pr > F
Model 1 27.49000 27.49000 18.00 0.0022
Error 9 13.74249 1.52694
Corrected Total 10 41.23249
Root MSE 1.23570 R-Square 0.6667
Dependent Mean 7.50091 Adj R-Sq 0.6297
Coeff Var 16.47394

Parameter Estimates
Parameter Standard
Variable DF Estimate Error t Value Pr > |t|
Intercept 1 3.00173 1.12392 2.67 0.0256
x 1 0.49991 0.11782 4.24 0.0022
 PA R T

Design
Principles
 CHAPTER

Principles of
Communication-
Effective
Graphic Design
The principles presented here are actually software independent, but the mission
of this book is to help you implement them with SAS ODS Graphics. Among the
numerous principles presented here, three key design objectives that deserve
special emphasis and should always guide your graphic design are as follows:

• Provide precise numbers whenever possible, rather than

force the graph viewer to estimate them, such as the Y
and X values for plot points. A graph enables quick, easy
inference, but the precise numbers are needed for correct
inference.
• Show the viewer what’s important.
• Assure readability. It is not automatic.

© LeRoy Bessler 2023

LeR. Bessler, Visual Data Insights Using SAS ODS Graphics,
https://doi.org/10.1007/978-1-4842-8609-8_1
4 Chapter 1 | Principles of Communication-Effective Graphic Design

Be Brief, Clear, Picturesque, and Accurate

I try to always be guided by the following quote from Joseph Pulitzer:

Put it before them briefly so they will read it,

clearly so they will appreciate it,
picturesquely so they will remember it,
and, above all, accurately so they will be guided by its light.

Though not about graphic communication, this quote by the famous newspaper
publisher (known for the Pulitzer Prizes, which were established in 1917 as a
result of his endowment to Columbia University) is wise advice for any
communication. A graph is inherently picturesque, but is not automatically
clear nor devoid of superfluous content. If the source data is accurate, the
graph will be an accurate representation of information—if its design does
not distort it.

3D Pie Charts Are Always Misleading

Look at Figure 1-1, an illuminating and convincing example that 3D pie charts
are always misleading (unless they consist of only two or four slices). The
slices are laid out in this fortuitous way only because the slices are really
ordered alphabetically by the slice names that I have chosen to omit. I did not
deliberately create this proof that 3D pie charts distort and falsely
communicate. The inherently dangerous 3D pie chart did it for me.

Figure 1-1. 3D pie chart versus 2D pie chart

 Visual Data Insights Using SAS ODS Graphics 5

3D Bar Charts Are Anticommunicative

The subtitles in the bar charts in Figure 1-2 explain how and why 3D bar
charts are anticommunicative. For the 2D example, I suppress the really not
very helpful X axis and display the precise values in a column adjacent to the
bars—a communication-effective graph needs to display precise numbers, not
just an image with axis values to estimate numbers.
ODS Graphics cannot create 3D bar charts, but there is the option to fill bars
with what are called “data skins” instead of uniform solid fill color. A data skin
can produce bars that look puffy, like long buttons, and can produce a puffy
pie. The data skin adds no communication value. It is nothing more than a
3D-like effect that can be used, if desired, for decoration.

Figure 1-2. 3D bar chart versus custom 2D bar chart

Graphs Need Image and Precise Numbers

Provide image plus precise numbers:

• Image for quick and easy inference

• Precise numbers for accurate and reliable inference

Consider the pair of bar charts that were presented in Figure 1-2. It is
impossible to reliably determine precise numbers by comparing bar ends (or
point locations for a scatter plot or trend line) to axis values. Moving your eye
from a bar end (or a plot point) to an axis and mentally interpolating the
approximate corresponding point on the axis to estimate a number based on
the framing tick mark values is a futile, unreliable, unacceptable way to try to
get the precise number and certainty.
6 Chapter 1 | Principles of Communication-Effective Graphic Design

In Figure 1-3 are two ways to present a scatter plot of data. In the right-hand-
side plot, there is no guesswork required to get the precise Y and X values.
Then in Figure 1-4, drop lines are used. They can get the viewer’s eye to the
correct place on the axis, but the viewer still needs to estimate, based on the
“enclosing” tick mark values. But drop lines are helpful when you do not have
annotation of the Y and X values. In Figure 1-4, annotation is instead used to
deliver information other than the scatter plot coordinates. That scatter plot
could have been made more informative by including Age in the data labels
and by color-coding the markers based on Sex which is a variable in the input
data set.

Figure 1-3. Scatter plot with grid and tick mark values versus annotated scatter plot
 Visual Data Insights Using SAS ODS Graphics 7

Figure 1-4. Using drop lines makes axis values and tick marks more helpful

When neither annotation nor drop lines are sufficient to get numbers for a
scatter plot, the only solution is a companion table, on the same page, or as
an Excel table that is linked to a web-deployed scatter plot.
A pie chart should include descriptions, values, and percents of the whole.
When labels overlap or (for tiny slices) disappear, use a legend instead of
labels as in Figure 1-5.
8 Chapter 1 | Principles of Communication-Effective Graphic Design

Figure 1-5. Fully informative ranked pie chart with legend

A legend requires moving the eyes from each slice to its legend entry. A good
alternative is the ODS Graphics CALLOUT option shown in Figure 1-6, where
there is a dashed line from each slice to its label. An extra benefit is avoiding
the preprocessing to prepare the fully informative legend entries.

Figure 1-6. Fully informative ranked pie chart with callout labels

A bar chart can have its numbers in a column next to its bars, as in the 2D bar
chart in Figure 1-2. Another way to provide those numbers, and more, is to
use Y axis tables as columns next to the bar category labels. In Figure 1-7,
 Visual Data Insights Using SAS ODS Graphics 9

three Y axis tables are used to create what is a ranked and rank-labeled bar
chart alternative to a pie chart.

Figure 1-7. A ranked order rank-labeled fully informative horizontal bar chart

For a large number of categories, you can provide them in alphabetical order
for easy lookup, but still can provide Rank as shown in Figure 1-8.

Figure 1-8. An alphabetical order rank-labeled fully informative horizontal bar chart

A Usable Stacked Bar Chart Requires

an Axis Table to Deliver the Precise Values
A usable stacked bar chart must have an axis table to deliver the precise
values. For Figure 1-9, the only thing you can do is visually compare the sizes
of bar segments, but with no accurate knowledge of the Sales values.
10 Chapter 1 | Principles of Communication-Effective Graphic Design

Figure 1-9. This typical stacked bar chart is unable to provide precise values of sales

In a bar chart with more segments, moving your eye from bar segments to the
X axis table is not easy or efficient, but Figure 1-10 is truly informative and
allows a viewer to quickly and easily identify Product-Region combinations
with larger sales.

Figure 1-10. Usable stacked bar chart with precise values for the bar segments

The red segments in two of the stacked bars in Figures 1-9 and 1-10 is a good
demonstration that color with insufficient mass is difficult to distinguish.
Color distinguishability is discussed further in Chapter 2.
 Visual Data Insights Using SAS ODS Graphics 11

 Clustered Horizontal Bar Chart Is

A
Better Than a Stacked Bar Chart
When there are two categorical key variables, Figure 1-11 is a better solution.
The precise values are adjacent to their bars.

Figure 1-11. A clustered horizontal bar chart

With no legend, the color is not needed for legend purposes, but it allows the
viewer to more quickly visually compare the sales in any region across all
products. Rather than have to find the bar label in the list of region bars for
each product.

For Bar Charts, Horizontal Is Usually Better

On a vertical bar chart, the bar labels along the X axis or the values that you
want to display at the bar ends might be too wide. On a horizontal bar chart,
space problems are less likely.

For a Line Plot with Discrete X Values, an X

Axis Table Is an Alternative to Annotation
An X axis table can be color-coded so that no legend is necessary (see
Figure 1-12). Instead of relying on going to look for a legend, the identity of
any plot point is already in the row label for Y values in the color that
corresponds to the color of the line. This color-coded plot is an efficient and
effective way to find all of the information needed to interpret a multi-line plot.
12 Chapter 1 | Principles of Communication-Effective Graphic Design

Figure 1-12. Multi-line overlay plot with X axis table

Furthermore, unlike the case in Figure 1-13, on a multi-line plot, even if the
plot points within a line are sufficiently separated to make annotation in prin-
ciple possible, collisions between annotation and other lines or between
annotation for one line with annotation for another line are always likely,
unless the lines happen to be well separated throughout their extent, which
is not the case in Figure 1-12.

Curve Labels Eliminate the Need for a Legend

On a multi-line plot, the effect of collisions between annotation and lines can
be mitigated by using a lighter shade for each line’s color. The curve labels in
Figure 1-13 provide an easier and faster way for a viewer to identify lines than
a legend.

Figure 1-13. Multi-line overlay plot with data labels and curve labels, no Y axis needed and
no legend needed
 Visual Data Insights Using SAS ODS Graphics 13

Show Them What’s Important

Suppressing superfluous graphic paraphernalia, as is discussed later in this
chapter (and demonstrated with coding in Chapter 3), limits what the viewers
look at and is consistent with the principle of Showing Them What’s Important.
But showing what’s important is very easy to implement for bar charts and pie
charts with ranking and for bar charts with subsetting. The discussion here
pertains to data with a categorical key, such as Sales by Region.

Show Them What’s Important with Ranking

Order the bars in a bar chart or slices in a pie chart by the value of the
measurement of interest, from largest to smallest. Show the Rank as part of
the bar or pie slice labeling.
If you have a chart where a small value is desirable, instead order the bars or
slices from smallest to largest.
In a case where the number of bars in a chart is large, such as the 50 states of
America plus the District of Columbia, you might instead order the bars
alphabetically for easy lookup. Perhaps better, you could present the bar chart
twice, both alphabetically and by magnitude of the measurement.

Show Them What’s Important with Subsetting

By subsetting I mean limit the data presented, rather than delivering all of it.
If you have ranked the data, this increases the focus on what’s important. For
example, instead of creating a bar chart of the population of every one of the
195 countries in the world, you might present a subset of only the Top 25,
Top 10, or Top Whatever. There are the familiar Top 10 lists. For music, there
are the Top 40 lists.
There is also the concept called the Pareto Principle, or “The 80-20 Rule.” It
says that in many situations, about 80% of the results come from 20% of the
causes. For example, 80% of an enterprise’s sales might come from 20% of its
customers. Since they have shown themselves to be more heavily interested,
that 20% might be the ones to be more frequently solicited for additional
purchases. Another variation of this phenomenon could be that 80% of an
enterprise’s problem reports might come from 20% of its customers. There is
nothing magical or inevitable about 80 and 20. The percentages might be, say,
90 and 10. The key concept, for graphic design, is that not every category is
equally important, and it is useful to focus on the most important for
enterprise operation.
14 Chapter 1 | Principles of Communication-Effective Graphic Design

Another way to think about subsetting is the idea of limiting the message.
Always remember Pulitzer’s advice from the beginning of this chapter: “Put it
before them briefly.” I once received the same advice from my assistant
Kenneth J. Wesley when I was agonizing over a report to executive
management. He said, “LeRoy, put it on one page. If you make it longer, they
won’t read it.”
Some time later, at the suggestion of Atis Purins, I was doing a design and
construction makeover for reports on the performance, capacity, and usage
of all of Miller Brewing Company’s computer resources. One report was on
disk capacity consumption and ran to numerous pages. By limiting the number
of consumption purpose categories ordered by size to one page, attention
was drawn to a much shorter list, and it accounted for a huge percent of the
total consumption burden, thus showing the information that was most
important. What had been unwieldy became readable, and when any new
report features were added, I was directed to adhere to the new design
concept. The decision to use subsetting and ranking limited the volume of
information, and kept the focus where appropriate.
When subsetting the input categories for a bar chart, it is essential to inform
the viewer as to the relative significance of what’s in versus what’s out. As
demonstrated in Figure 1-14, use the title and subtitles to give the viewer of
the graph these facts:

• The total number of categories

• The grand total of the measurement over all categories
• The number of categories shown
• The subtotal of the measurement for the categories shown
• What percent of the grand total is that subtotal

Three Ways to Do Ranked Subsetting

An extremely popular way, anywhere, to do ranking and subsetting is with, for
example, a Top Ten List. From this first way that I thought of, the title of a
report was “Top 20 Cities Account for Subtotal Shoe Sales of $24,402,060
which is 72.09% of the Grand Total.” And in that case, the subtitle was “All 53
Cities Have Grand Total Shoes Sales $33,851,566.” That pair of statements
embodies the design and information principles for titles that I presented in
the preceding section.
A second obvious solution is to use a cutoff, in which case a possible title
could be something like “Ranked X Cities with Sales of At Least $1,000,000
Account for…,” where X would be the number that pass the test and where
the end of the title would be analogous to that of the Top 20 Cities report.
 Visual Data Insights Using SAS ODS Graphics 15

A third way of subsetting that I particularly like is to limit the ranked categories
to only enough to account for a specified percent of the grand total of the
measurement of interest, as in Enough Ranked Cities to Account for At Least
90% of Grand Total Shoe Sales. This avoids picking an arbitrary N for Top N
or an arbitrary cutoff for the measure of interest (see Figure 1-14).
In Chapter 4, all three ways to do subsetting are shown in Figure 4-7.

Figure 1-14. Ranked and subsetted bar chart, selecting subtotal percent of grand total

If you like, you could web-deploy multiple subsets and allow the viewer to
navigate among them with links. You could even include a link to a complete
list of all of the categories. How to web-deploy interlinked graphs with ODS
HTML5 will be shown in Chapter 14.

S implicity Accelerates and Facilitates

Visual Insights into Data
Simplicity accelerates and facilitates visual insights into data. Needless
complexity impedes and/or obstructs communication. Simplicity is powerful.
Simplicity is elegant.
Elegance (like in an elegant mathematical proof) consists of

• Everything necessary
• Only the necessary
16 Chapter 1 | Principles of Communication-Effective Graphic Design

Let’s take a look at the simplest graph I’ve ever created. At first glance, the pie
charts in Figure 1-15 look trivial, but they really are not. I usually advocate
against consolidating very small slices in a pie chart or very small bars in a bar
chart into an “Other” category. It prompts the question, “What is in ‘Other’?”
Your graph design should anticipate and answer all questions, not create them.

Figure 1-15. The Extremes of Other with “The Pac-Man Pie Chart”

Here, the smallness of “Other” in the chart on the left dramatizes the fact
that the sum of all of the competitive products’ market shares was insignificant.
The sizes and values for their tiny market shares were deemed by me as not
worth showing.
Conversely, the chart on the right is expressly meant to prompt interest in
what is in “Other.” I created it during my time as a local elected official, to
emphasize to residents that the bulk of their real estate property tax payment
went to other public bodies which were also getting a share (in some cases, a
much larger share). Since the tax was paid to the office of our Village Treasurer,
he would receive letters of concern about the size of the bill. To anticipate
and address questions and concerns, the pie chart was included in a cover
letter that went out with the annual tax bill. There also was a table of
supporting detail, which compared the current amounts and growth of all of
the shares of tax bill total. We did explain what was in “Other,” but after
graphically emphasizing the smallness of Village government’s share of the bill.
I call this design “The Pac-Man Pie Chart,” and I call these examples “The
Extremes of Other.” Why “Pac-Man”? Who is that guy? Pac-Man is a video
game that was introduced in 1980. It features a little creature who looks like
a yellow pie with a slice missing.

A Sparse Graph Is Easily and Quickly Interpreted

Two examples of this arose back in 1990, when I developed a way of sometimes
presenting trend data (also known as “time series” data) that I called “Sparse
Line Annotation.” Of course, either Pac-Man pie chart in Figure 1-15 also
qualifies as a “sparse graph” because of its simplicity.
 Visual Data Insights Using SAS ODS Graphics 17

I was working at Miller Brewing Company and wanted to graphically compare

two historical data sets, one for Miller Lite and the other for all beer in
America. For both plots, knowing every precise Y value was not important.
The interesting feature is the behavior of each trend and an apparent correlation
in their behaviors. However, that correlation is not the focus of this discussion.
What is sparse on this pair of graphs is my choice of which Y values to annotate
and which X values to show on the axis. In later chapters, you will see a more
compact and even sparser way to present the essential data for a trend line.
From Figure 1-16, I concluded that the most interesting data in a trend might
usually be the values of Y (the measure of interest) and X (the date) for

• The start
• The end
• Any intermediate maximum
• Any intermediate minimum
• The size and direction of change since the date previous
to the end date (not shown in the examples created then
and presented here)

Figure 1-16. Sparse Line Annotation case 1: start, end, and maximum

From Figure 1-17, I realized that an additional point of interest could be a

point, not a minimum or maximum, where the trend permanently changes in
any of these ways:

• From rapidly increasing to slowly increasing

• From slowly increasing to rapidly increasing
18 Chapter 1 | Principles of Communication-Effective Graphic Design

• From rapidly decreasing to slowly decreasing

• From slowly decreasing to rapidly decreasing

Figure 1-17. Sparse Line Annotation case 2: start, end, and trend change

Sparse Line Annotation has been updated and significantly enhanced as is

shown in Figure 1-18, which is taken up as Figure 8-9. As an alternative to an
overlay plot, see also Sparse Line Tables and Sparse Line Panels in Figures 9-9,
9-11, 9-12, and 9-13 and the web-enabled Sparse Line Table in Figure 14-15.
The new Sparse Lines remove the X axis entirely and provide X values as part
of the data point annotation, not just the Y values. An important new feature
in the new Sparse Line, standalone or in any of the composites, is to provide,
as additional annotation for the last data point, the change since the second
last data point, which is always of interest when looking at the last data point.

Figure 1-18. The updated and enhanced Sparse Line, using different data
 Visual Data Insights Using SAS ODS Graphics 19

Inform the Viewer About the Key Data Points

of a Multi-line Overlay Plot
As I previously explained, the most interesting Y values on any trend line are
usually the start, end, and any intermediate minimum or maximum. Picking
those out visually is easy on a single-line plot or when the number of X values
is small on a multi-line plot. A kinder alternative for any multi-line overlay plot
is to deliver the information to the viewer, rather than require it to be
discovered by scrutiny. The viewer of your output will appreciate convenience,
completeness, and certainty in the information delivery.

Figure 1-19. Providing all of the essential information for an overlay plot when annotation is
infeasible
20 Chapter 1 | Principles of Communication-Effective Graphic Design

For a Trend, Usually Start the Y Axis at Zero

In Figure 1-16, I started the Y axis at zero. That happened to make the trend
line for the first plot look nearly flat. It is interesting and appropriate that
industry observers described this range of American beer consumption
history as flat.
SAS ODS Graphics, like other software, by default will take the range of the
Y variable and allow it to use all of the vertical space. If you have a need or
desire to see the trend with maximum variation, that is the right design.
However, the disadvantage is that it can cause a trend viewer unnecessary
anxiety about unfavorable slope in the trend or unjustified elation about
favorable slope. Rather than react to the most recent slope of the trend, the
latest value on a trend line should be evaluated based on the magnitude of the
change or based on whether or not the latest value has reached a desired goal
or crossed a dangerous threshold. If there is a desired goal or a dangerous
threshold, include a reference line for it on the graph.
If you use web delivery for the trend line, you can present both views of the
trend hyperlinked together, to provide both an unexciting view of the trend
and the ability to see it with the maximum variation. An alternative option is
to stack the two views in a document page.
The recommendation to start the Y axis at zero is always practical for a sin-
gle-line plot. However, for an overlay plot, it can be impractical. For example,
even if the table of information at the bottom of Figure 1-19 were omitted,
the already crowded and crisscrossing plot lines would be squashed together
in a useless mess.

F or a Bar Chart, Unless There Are Negative

Values, Always Start the Value Axis at Zero
This avoids distorting apparent relative magnitude of the bars. From time to
time, I see the unfortunate decision to start a bar chart somewhere above
zero. Presumably, the reason is to save vertical space, but the result is
deceptive, intentionally or not.

 se Maximally Simple Design

U
to Focus on What’s Important
We can put a graph “before them briefly” and maximize focus where it’s
needed by keeping the layout simple as possible. Here, my concern is not
limiting the information or the graphic elements, but on omitting and
suppressing non-information and superfluous graphic artifacts.
 Visual Data Insights Using SAS ODS Graphics 21

Before computer graphics software became available, we manually plotted

our data with a pen or pencil on graph paper. If you have never seen it, you
can find pictures online. Graph paper was usually a grid of green lines on white
paper. ODS Graphics can emulate it. That is, as shown in the left-hand-side
plot in Figure 1-3, it can provide grid lines, axis lines, tick marks, minor tick
marks, axis values, and a border around the plot area. In the right-hand-side
plot in Figure 1-3, I suppressed the traditional graphic framework and added
annotation for the plot points, using ODS Graphics data labels.
Feasibility of annotation of a scatter plot depends on plot point density. With
fewer, I could have included the names of the people. The annotated plot
delivers the information briefly (i.e., without superfluous graphic paraphernalia),
clearly, picturesquely, and accurately—there is no guesswork required to get
the precise Y and X values. With annotation, the viewer gains information,
and the image loses the superfluous axis lines, axis values, axis tick marks, and
grid lines. In Chapter 3, you will see, among other things, the code used to
create the annotated scatter plot with its maximally simple framework.
Another example of suppressing the superfluous graphic paraphernalia is the
2D bar chart in Figure 1-2. It briefly delivers the bars, the categories, and the
values, the only things that the viewer needs to see and needs to know. There
are no axes, no axis labels, no tick marks, and no tick mark values—which
would need to be used to guess the precise values for each category if the
column of values had not been provided instead.
I saw a recommendation to display the minimum of traditional graphics
paraphernalia in Edward Tufte’s presentation at a Computer Measurement
Group conference in the 1980s.

Tell Them What’s Important with a Headline

The typical title of a graph tells the viewer what is being shown. If that is not
self-evident from the graph, such information can instead be put in the second
title line when devoting the first line to a headline. If there is something
especially significant in the graph, or some inference that can be drawn from
the graph, or the graph shows a very encouraging result or implicitly carries a
cautionary message, do not leave it to the viewer to discover it. Just come
right out and SAY IT, in a headline.

A Graph Footnote Does Not Need to Be Small

If the information is important, do not undersize it. A footnote that is merely
identifying a data source, or providing some boilerplate statement(s), could
reasonably be in a small font size, to save vertical display space and to avoid
distraction from important information or the image.
22 Chapter 1 | Principles of Communication-Effective Graphic Design

When the footnote contains important information, make it the same size as
the title. ODS Graphics title lines are always Bold by default. Consider making
your footnote(s) Bold as a standard, unless a footnote is better diminished for
cases mentioned in the preceding paragraph.

Assure Text Readability

This seems like an obvious design guideline. Readability is always important,
but it is not automatic. You should not assume it.
Text readability is essential since the letters and numbers on a graph are as
important as the graphic elements. Image accelerates and facilitates inference
from, and understanding of, the data, but precise data (the labels and numbers)
assure reliable and correct inference and understanding. I emphasized that at
the opening of this chapter. In Chapter 2, we will see that color “readability”
also matters, whether it is color of text or color of graphic elements.
If you are satisfied with how the software defaults format the text for a
specific graph without your specifying text characteristics, that’s the simplest
solution. If not, adjust the software’s options to deliver text and numbers you
deem most readable.

Font Size and Font Weight Affect Readability

The main determinants of readability are font size and font weight. Thicker
letters and numbers (i.e., those with Bold weight) are always more readable.
Despite best intentions, actual font size used might sometimes have to be
dictated by getting text to fit without overlap of adjacent text items, such as
plot point annotations. In that case, sparse annotation might be a useful
workaround. In the situations demonstrated in Figures 1-16 and 1-17, sparse
annotation is a benefit, not a deprivation of important information.
It is believed that sans serif fonts, such as Arial, are better for visually impaired
readers. If sans serif fonts benefit visually impaired readers, then I contend
that they can benefit all readers. An advantage of sans serif fonts is that they
are drawn with a uniform thickness throughout the contour of the character,
rather than with thin and thick points around their contour. Thick is always
easier to see. There is a counterargument in favor of serif fonts, but it pertains
to reading large amounts of text, such as in a novel. For a graph, serif fonts are
not an advantage for any viewer.
 Visual Data Insights Using SAS ODS Graphics 23

Keep Text Horizontal

Use horizontal text whenever possible. That is the way that we read most
languages. One often sees vertical text used as the label for a vertical axis.
Forcing it to be horizontal might force the displayed graph area to be narrower.
A simple alternative is to put the axis labels in a subtitle, if they are not already
in the graph title, either explicitly or implicitly.
If the horizontal axis is obviously the date, there is no value in labeling it. Use
the space available to maximize the vertical display space for the graph. This
is a more efficient use of the graph viewer’s time and attention. Tell them
what’s important, not what they can know at a glance, without explicitly
being told.
Another place where one sometimes sees vertical text or tilted text is as the
tick mark values on the horizontal axis. When some horizontal axis tick mark
values are automatically omitted by software (in ODS Graphics, this is called
“thinning”), if the omitted ones can be inferred from adjacent displayed values
to the left and the right of where there would be a tick mark value, then no
harm is done. Apart from options to control the characteristics of the tick
mark values (font face, font size, font weight, font color), ODS Graphics
provides controls on thinning, as will be shown later. There is also an option
to use STAGGER on the X values, by displaying them in two rows below the
X axis. As you move from left to right, the X value is in the top row, then in
the bottom, then back in the top, and so on. If you cannot turn off the thinning
with an option, and thinning causes a problem, then reducing font size is the
obvious solution.

Never Use Backgrounds—They Impair Readability

The readability of text is too often diminished or even obstructed by
decorative features such as textured backgrounds, color gradient backgrounds,
and, what can easily be the worst, photographs. The graphic part of your
image should not have to compete for the attention of the viewer. The graph
and the text in your ODS Graphic image needs a plain solid color background,
preferably white, but in any case definitely a background color of high contrast
with the text color.
The use of colored text or text on colored backgrounds is discussed in
Chapter 2. The simplest and most important principle for color is: use color
to communicate, not to decorate.
24 Chapter 1 | Principles of Communication-Effective Graphic Design

Readability Depends on Display Situation

If your audience will receive hardcopy of your graph or might print your
graph, be sure to evaluate your graphic output as printed. Print it! Verify that
what you get is what you expected, what you intended. The image and text
on your monitor screen are shining light at your eyes. Hardcopy is viewed
with reflected light which presents a weaker image to your eyes, with the
consequence that text, thin lines, small plot point markers, and legend color
swatches can be difficult to read or to interpret as to precise color.
If creating color output that is expected to be printed by somebody, try to
assure that they have access to a color printer.

A Graph Can Have a Companion Table

When you cannot provide the precise numbers inside the graphic image, a
companion table is the solution. You can deliver a graph and its companion
table on the same page, or on separate pages in the same file, if delivering as
a PDF or Microsoft Word document. In Excel, you can deliver a graph and its
companion table on the same worksheet or in separate worksheets in the
same workbook. In HTML, they can be on the same web page or on interlinked
web pages.

Web Graphs
There are special considerations and capability to consider when presenting
your graphs with HTML packaging.

Include Data Tips (a.k.a. Mouseover Text)

Whenever permanent annotation does not fit, include data tips via mouseover
text. This displays the precise numbers, but, of course, they disappear when
the mouse moves on. Besides displaying the values being graphed, you can
optionally display additional information related to the graphic element as in
Figure 1-20.
 Visual Data Insights Using SAS ODS Graphics 25

Figure 1-20. Web plot with maximally informative data tips

 Web Graph Can Have a Companion

A
Hyperlinked Excel Table
Consider providing the web graph viewer the opportunity to inspect, using
Excel, all of the data statically (unlike data tips, which are temporary). This is
a good solution for a scatter plot or a trend line that is too dense for
annotation to provide precise numbers. (You can instead put the table on the
same or another web page.)
A recipient of your web graph is likely to have Excel on their computer and
the knowledge of how to use it. How to deliver a web graph with a linked
Excel table companion will be explained later. The Excel table can be
hyperlinked back to the web graph.
You can deliver not just the values of the variables being graphed, but any
additional data elements that might be in the input data set used for the graph.
26 Chapter 1 | Principles of Communication-Effective Graphic Design

A Web Graph Should Not Require Scrolling

A trend line, or any graph, is best viewed in its entirety. It must never require
horizontal scrolling. A horizontal bar chart with very many bars, if designed as
shown earlier, that does not require comparing bars with an X axis could be
usefully viewed with vertical scrolling.
Optimize the size of the image file to fill, but not exceed, what you expect to
be the probable size and shape of the screens used to view the web page—
Show Them the Entire Picture. If in doubt as to the best size, it would be
better to err in the direction of smaller, within the limits of readability. If
necessary, the web page viewer should be able to use the browser’s zoom
feature to enlarge or shrink the displayed material.

Summary
This is a review of the Principles of Communication-Effective Graphic Design.
These reminders are what I like to call my Principia Graphika, a title inspired
by Principia Mathematica by Sir Isaac Newton (1687). That title was reused for
three volumes by Alfred North Whitehead and Bertrand Russell (1910,
1912, 1913):

• Effective communication is brief, clear, picturesque, and

accurate so that it is read, understood, remembered, and
reliable.
• Provide image plus precise numbers: image for quick and
easy inference and precise numbers for accurate and
correct inference. Estimates based on axis tick mark
values are unreliable.
• A scatter plot or a trend (a.k.a. time series) plot, if not
too dense, can be annotated.
• For plots which cannot be annotated, an X axis table
might be able to deliver the Y values.
• For a plot where neither annotation nor an X axis table
is feasible, drop lines can show the correct locations on
the axes, but the viewer of your plot will still need to
estimate the value.
• On a multi-line plot, if full annotation is infeasible due to
label-label or label-line collisions, a color-coded X axis
table can deliver the Y values and can identify each plot
line without a legend.
Random documents with unrelated
content Scribd suggests to you:
 It was a Greek pedant who stood before a mirror and shut his
eyes that he might know how he looked when asleep—a jest which
reappears in Taylor’s Wit and Mirth in this form: “A wealthy monsieur
in France (hauing profound reuenues and a shallow braine) was told
by his man that he did continually gape in his sleepe, at which he
was angry with his man, saying he would not belieue it. His man
verified it to be true; his master said that he would neuer belieue any
that told him so, except (quoth hee) I chance to see it with mine
owne eyes; and therefore I will have a great Looking glasse at my
bed’s feet for the purpose to try whether thou art a lying knaue or
not.”
Not unlike some of our “Joe Millers” is the following: A citizen of
Cumæ, on an ass, passed by an orchard, and seeing a branch of a
fig-tree loaded with delicious fruit, he laid hold of it, but the ass went
on, leaving him suspended. Just then the gardener came up, and
asked him what he did there. The man replied, “I fell off the ass.”—
An analogue to this drollery is found in an Indian story-book, entitled
Kathȧ Manjari: One day a thief climbed up a cocoanut tree in a
garden to steal the fruit. The gardener heard the noise, and while he
was running from his house, giving the alarm, the thief hastily
descended from the tree. “Why were you up that tree?” asked the
gardener. The thief replied, “My brother, I went up to gather grass for
my calf.” “Ha! ha! is there grass, then, on a cocoanut tree?” said the
gardener. “No,” quoth the thief; “but I did not know; therefore I came
down again.”—And we have a variant of this in the Turkish jest of the
fellow who went into a garden and pulled up carrots, turnips, and
other kinds of vegetables, some of which he put into a sack, and
some into his bosom. The gardener, coming suddenly on the spot,
laid hold of him, and said, “What are you seeking here?” The
simpleton replied, “For some days past a great wind has been
blowing, and that wind blew me hither.” “But who pulled up these
vegetables?” “As the wind blew very violently, it cast me here and
there; and whatever I laid hold of in the hope of saving myself
remained in my hands.” “Ah,” said the gardener, “but who filled this
sack with them?” “Well, that is the very question I was about to ask
myself when you came up.”
 The Greek Anthology brings together short poems and epigrams
written during the thousand years between Simonides’ time and the
sixth century a.d.
Collected shortly before the beginning of the Christian Era and
added to later, they comprise about four thousand five hundred
specimens, by three hundred authors. Few of these are witty, as,
indeed, few are epigrammatic, but of them we quote some which
seem most appurtenant.

FROM THE GREEK ANTHOLOGY

Lucian
DARKNESS
“A blockhead bit by fleas put out the light,
And, chuckling, cried, ‘Now you can’t see to bite!’”

Crates
CURES FOR LOVE
“Hunger, perhaps, may cure your love,
Or time your passion greatly alter;
If both should unsuccessful prove,
I strongly recommend a halter.”

Julian
BEER
“What! whence this, Bacchus? For, by Bacchus’ self,
The son of Jove, I know not this strange elf.
The other smells like nectar; but thou here
Like the he-goat. Those wretched Celts, I fear,
For want of grapes, made thee of ears of corn.
Demetrius art thou, of Demeter born,
Not Bacchus, Dionysus, nor yet wine—
Those names but fit the products of the vine;
Beer thou mayst be from barley; or, that failing,
We’ll call thee ale, for thou wilt keep us ailing.”

Agathias
GRAMMAR AND MEDICINE
“A thriving doctor sent his son to school
To gain some knowledge, should he prove no fool;
But took him soon away with little warning,
On finding out the lesson he was learning—
How great Pelides’s wrath, in Homer’s rime,
Sent many souls to Hades ere their time.
‘No need for this my boy should hither come;
That lesson he can better learn at home;
For I myself, now, I make bold to say,
Sent many souls to Hades ere their day,
Nor e’er found want of grammar stop my way.’”

Nearchus
A SINGER
“Men die when the night-raven sings or cries;
But when Dick sings, e’en the night-raven dies.”

Ammianus
AN EPITAPH
“Light lie the earth, Nearchus, on thy clay,
That so the dogs may easier find their prey.”

Lucilius
ENVY
“Poor Diophon of envy died,
His brother thief to see
Nailed next to him and crucified
Upon a higher tree.”

A PROFESSOR WITH A SMALL CLASS

“Hail, Aristides, rhetoric’s great professor!
Of wondrous words we own thee the possessor.
Hail ye, his pupils seven, that mutely hear him—
His room’s four walls, and the three benches near him.”

FALSE CHARMS
“Chloe, those locks of raven hair,
Some people say you dye with black;
But that’s a libel, I can swear,
For I know where you buy them black.”
 A SCHOOLMASTER WITH A GAY WIFE
“You in your school forever flog and flay us,
Teaching what Paris did to Menelaus;
But all the while, within your private dwelling,
There’s many a Paris courting of your Helen.”

BOARD OR LODGING
“Asclepiades, the miser, in his house
Espied one day, to his surprise, a mouse.
‘Tell me, dear mouse,’ he cried, ‘to what cause is it
I owe this pleasant but unlooked-for visit?’
The mouse said, smiling, ‘Fear not for your hoard;
I come, my friend, to lodge, and not to board.’”

Anon
CONVENIENT PARTNERSHIP

“Damon, who plied the undertaker’s trade,

With Doctor Crateas an agreement made.
What linens Damon from the dead could seize,
He to the doctor sent for bandages;
While the good doctor, here no promise-breaker,
Sent all his patients to the undertaker.”

Anon
LONG AND SHORT

“Dick cannot blow his nose whene’er he pleases

His nose so long is, and his arm so short;
Nor ever cries, ‘God bless me!’ when he sneezes—
He cannot hear so distant a report.”

Anon
THE LERNEANS

“Lerneans are bad: not some bad and some not

But all; there’s not a Lernean in the lot,
Save Procles, that you could a good man call.
But Procles—is a Lernean, after all.”

Anon
PERPLEXITY

“Sad Heraclitus, with thy tears return;

Life more than ever gives us cause to mourn.
Democritus, dear droll, revisit earth;
Life more than ever gives us cause for mirth.
Between you both I stand in thoughtful pother,
How I should weep with one, how laugh with t’other.”

Beside his short poems, we quote a little of the prose of

Lucian
A QUESTION OF PRECEDENCE
zeus, æsculapius, and heracles

“Zeus. Do, Æsculapius and Heracles, stop your wrangling, in

which you indulge as if you were a couple of mortals; for this sort of
behavior is unseemly, and quite strange to the banquets of the gods.
“Heracles. But, Zeus, would you have that quack drug-dealer
there take his place at table above me?
“Æsculapius. By Zeus, yes, for I am certainly the better man.
“Heracles. How, you thunderstruck fellow, is it, pray, because
Zeus knocked you on the head with his bolt for your unlawful actions,
and because now, out of mere pity, by way of compensation, you
have got a share of immortality?
“Æsculapius. What! have you, for your part, Heracles, altogether
forgotten your having been burned to ashes on Mount Œta, that you
throw in my teeth this fire you talk of?
“Heracles. We have not lived at all an equal or similar sort of life—
I, who am the son of Zeus, and have undergone so many and great
labors, purifying human life, contending against and conquering wild
beasts, and punishing insolent and injurious men; whereas you are a
paltry herb-doctor and mountebank, skilful, possibly, in palming off
your miserable drugs upon sick fools, but who have never given
proof of any noble, manly disposition.
“Æsculapius. You say well, seeing I healed your burns when you
came up but now half-burned, with your body all marred and
destroyed by the double cause of your death—the poisoned shirt,
and afterward the fire. Now I, if I have done nothing else, at least
have neither worked like a slave, as you have, nor have I carded
wool in Lydia, dressed in a fine purple gown; nor have I been beaten
by that Omphale of yours, with her golden slipper. No, nor did I, in a
mad fit, kill my children and my wife!
“Heracles. If you don’t stop your ribald abuse of me at once, you
shall very speedily learn your immortality will not avail you much; for
I will take and pitch you head first out of heaven, so that not even the
wonderful Pæon himself shall cure you and your broken skull.
“Zeus. Have done, I say, and don’t disturb the harmony of the
company, or I will pack both of you off from the supper-room;
although, to speak the truth, Heracles, it is fair and reasonable
Æsculapius should have precedence of you at table, inasmuch as he
even took precedence of you in death.”
—“Dialogues of the Gods.”

ODYSSEUS’S TRICK ON POLYPHEMUS

Poseidon and Polyphemus

“Polyphemus. Oh, father, what have I endured at the hands of the

cursed stranger, who made me drunk and put out my eye, assaulting
me when I was lulled to sleep!
“Poseidon. Who dared to do this, my poor Polyphemus?
“Polyphemus. In the first instance, he called himself Outis; but
when he had got clear away, and was out of reach of my arrow, he
said that his name was Odysseus.
“Poseidon. I know whom you speak of—him of Ithaca, and he was
on his return voyage from Ilium. But how did he do it, for he is by no
means a man of too much courage?
“Polyphemus. Returning from my accustomed attending of my
flock, I caught a number of fellows in my cave, evidently having
designs on my herds; for when I placed the stone block against the
door—the rock is of huge size—and had lighted the fire by igniting
the tree which I brought from the mountain, evidently they appeared
to be trying to conceal themselves. Well, when I had got hold of
some of them I devoured them for a pack of thieves, as was
reasonable. Hereupon that most villainous rascal, whether he was
Outis or Odysseus, pours out a sort of drug and gives me to drink—
sweet, indeed, and of delicious smell, but most insidious, and which
caused great disorder in my head; for, immediately upon my
drinking, everything seemed to me to be in a whirl, and the cave
itself was turned upside down, and I was no longer at all in my
senses; and at last I was dragged down into sleep. Then sharpening
the bar, and igniting it besides, he blinded me as I slept, and from
that time I am a blind man, at your service, Poseidon.
“Poseidon. How soundly you slept, my son, that you did not jump
up while you were being blinded! But as for this Odysseus, then, how
did he escape? For he could not—I am well assured that he could
not—move away the rock from the door.
“Polyphemus. Yes, but it was I who removed it, that I might the
better catch him as he was going out; and, sitting down close to the
door, I groped for him with extended hands, letting only my sheep go
out to pasture, after having given instructions to the ram what he was
to do in my place.
“Poseidon. I perceive: they slipped away unnoticed, under the
sheep. But you ought to have shouted, and called the rest of the
Cyclopes to your aid.
“Polyphemus. I did summon them, father, and they came. But
when they asked the sneaking rascal’s name, and I said it was Outis,
thinking I was in a mad fit, they took themselves off at once. Thus
the cursed fellow tricked me with his name; and what especially
vexes me is, that he actually threw my misfortune in my teeth. ‘Not
even,’ said he, ‘will your father Poseidon cure you.’
“Poseidon. Never mind, my child, for I will revenge myself upon
him; he shall learn that, even if it is not possible for me to heal the
mutilation of people’s eyes, at all events the fate of voyagers is in my
hands. And he is still at sea.”
—Dialogues of the Sea-Gods.

Remembering that the dividing lines may not be too strictly drawn,
we close our survey of Greek Humor with some of the fragments of
Menander.
Menander, who was to the Middle or New Comedy what
Aristophanes was to the Old Comedy, left only fragments. One bit,
rather longer than the others, shows, with the inevitable animal
element not lacking, a surprisingly modern spirit of satire.
“Suppose some god should say: Die when thou wilt,
Mortal, expect another life on earth;
And for that life make choice of all creation
What thou wilt be—dog, sheep, goat, man, or horse;
For live again thou must; it is thy fate;
Choose only in what form; there thou art free.
So help me, Crato, I would fairly answer
Let me be all things, anything but man.
He only of all creatures feels afflictions.
The generous horse is valued for his worth.
And dog by merit is preferred to dog,
And warrior cock is pampered for his courage,
And awes the baser brood. But what is man?
Truth, virtue, valour, how do they avail him?
Of this world’s good the first and greatest share
Is flattery’s prize. The informer takes the next.
And barefaced knavery garbles what is left.
I’d rather be an ass than what I am
And see these villains lord it o’er their betters.”

Other Fragments of Menander follow.

“Be off! these shams of golden tresses spare;
No honest woman ever dyes her hair.”

“Better to have, if good you rightly measure,

Little with joy than much that brings not pleasure,
Scant means with peace than piles of anxious treasure.”

“Marriage, if truth be told (of this be sure),

An evil is—but one we must endure.”

“Wretched is he that has one son; or, rather,

More wretched he who of more sons is father.”

“Think this, on marriage when your mind is set:

 If the harm is small, ’tis the chief good you’ll get.”

“Slave not for one who has been himself a slave;

Steers, loosed from ploughs, of toil small memory have.”

“A handsome person, with perverted will,

Is a fine craft that’s handled without skill.”

“Let not a friend your cherished secrets hear;

Then, if you quarrel, you’ve no cause for fear.”

“More love a mother than a father shows:

He thinks this is his son; she only knows.”

“Fathers’ and lovers’ threats no truth have got.

They swear dire vengeance,—but they mean it not.”

“Your petty tyrant’s insolence I hate;

If wrong is done me, be it from the great.”

“A lie has often, I have known before,

More weight than truth, and people trust it more.”

“Don’t talk of birth and family; all of those

Who have no natural worth on that repose.
Blue blood, grand pedigree, illustrious sires
He boasts of, who to nothing more aspires.
What use long ancestry your pride to call?
One must have had them to be born at all!
And those who have no pedigree to show,
Or who their grandsires were but scantly know.”

“From change of homes or lack of friends at need,

And so have lost all record of their breed,
Are not more “low-born” than your men of blood;
A nigger’s well-born, if he makes for good!”

The following are a few more epigrammatic bits from the writings
of less noted contemporaries.
Philippides
’Tis easy, while at meals you take your fill,
To say to sickly people, Don’t be ill!
Easy to blame bad boxing at a fight,
But not so for oneself to do it right.
Action is one thing, talk another quite.

Your fortune differs as to bed and board;

Your wife—if ugly—can good fare afford.

DIPHILUS
Learn, mortal, learn thy natural ills to bear:
These, these alone thou must endure; but spare
A heavier load upon thyself to bring
By burdens that from thine own follies spring.

When I am asked by some rich man to dine,

I mark not if the walls and roofs are fine,
Nor if the vases such as Corinth prizes,—
But solely how the smoke from cooking rises.
If dense it runs up in a column straight,
With fluttering heart the dinner-hour I wait.
If, thin and scant, the smoke-puffs sideway steal,
Then I forebode a thin and scanty meal.

So plain is she, her father shuns the sight:

She holds out bread; no dog will take a bite.
So dark is she, that entering a room
Night seems to follow her, and all is gloom.

Apollodorus
Sweet is a life apart from toil and care;
Blessed lot, with others such repose to share!
But if with beasts and apes you have to do,
Why, you must play the brute and monkey too!

In youth I felt for the untimely doom

Of offspring carried to an early tomb.
But now I weep when old men’s death I see;
That moved my pity; this comes home to me.

Seek not, my son, an old man’s ways to spurn;

To these in old age you yourself will turn.
Herein we fathers lose a point you gain;
When you of “father’s cruelty” complain,
“You once were young,” we tauntingly are told.
We can’t retort, “My son, you once were old.”
 PART II
ROME
The Roman Juvenal observed, “All Greece is a comedian.” But he
could not say the same of his own country.
Though there was Roman Comedy and Roman Satire, the real
and spontaneous spirit of fun was conspicuously lacking in the tastes
and tendencies of the Romans.
Glory is attributed to Greece and grandeur to Rome, and it may
be the “sudden glory” of humor was an integral part of the Grecian
nature.
Yet we must not differentiate too carefully between the two, for the
literature of Greece and Rome is so fused and intermingled that only
a historian may take up the chronological tabulation.
For our purpose it is well to let the literature of the two countries
merge and continue the consideration of classic comedy without
over cautious regard for dates.
The Greek influence on literature of all ages will never disappear,
but the Greek spirit of pure joy and gaiety will, probably never
reappear.
From the beginnings of Greece, on through the existence of
Rome, and down through the Mediæval Ages, the world of letters
was self-contained, a single proposition. From 500 b.c. to 1300 a.d.
the traditions of primal Greece and Rome continued to be the
common possession of all Europe.
After that, literature became diverse and divergent among the
countries. It was independent as well as interdependent, but this
condition makes an inevitable division of time.
Greece, Rome, Mediæval Times,—these are the three sections of
the Middle portion of this book.
 Rome, then, considered by herself, brought forth little quotable
humorous literature, and what we have to choose from is ponderous
and heavy.
Like Greece, the first germs of Roman comic literature may be
traced to the religious festivals, which were marked by an admixture
of religious rites and riotous Bacchanalian orgies, where as the
crowds danced and sang and feasted, they became first hilarious
and then abusive and indecent.
Like the Greeks, the Romans used grotesque masks, large
enough to represent face and hair, too, the duplicates of which we
see decorating our theater proscenium arches and drop curtains to
this day.
It would seem these masks were universally made use of in their
dramatic performances, for all caricatures and grotesque drawings
show them.
In the burlesque entertainments there was a Buffoon,
corresponding to our clown, called a Sannio, from the Greek word
meaning a fool.
Later, undoubtedly, the Court Fool and the King’s Jester were the
natural successors of this character.
In all these masks the features were exaggerated and made
monstrous of form and size. But one reason for the greatly enlarged
mouth is that it was so shaped in order to form a sort of speaking
trumpet, that the actors’ voices might be heard at greater distance.
In contrast to the grotesquerie of enlargement, there was also a
branch of caricature which depicted the pigmies.
The legend of the pigmies and cranes is as ancient, at least, as
Homer, and many examples are found in the buried cities of
Herculaneum and Pompeii.
Comic Literature was not plentiful in the days of Early Rome. Up
to the second century b.c. we can glean but the two names, Plautus
and Terence.
 These two, nearly contemporary, founded their plays on the
comedies of Menander and a few other earlier dramatic writers.
Perhaps twenty plays are left us from the hands of these two
Romans, and these, though pronounced amusing by scholars who
can read the original text, are not what the modern layman deems
very humorous.
A few examples of them will suffice.

Plautus
MILITARY SWAGGER
pyrgopolinices, artotrogus, and soldiers

Pyrgopolinices. Take care that the luster of my shield is more

bright than the rays of the sun when the sky is clear, that, when
occasion comes, the battle being joined, ’mid the fierce ranks right
opposite it may dazzle the eyesight of the enemy. But I must console
this saber of mine, that it may not lament nor be downcast in spirits,
because I have thus long been wearing it keeping holiday, though it
so dreadfully longs to make havoc of the enemy. But where is
Artotrogus?
Artotrogus. Here he is; he stands close by the hero, valiant and
successful, and of princely form. Mars could not dare to style himself
so great a warrior, nor compare his prowess with yours.
Pyrgopolinices. Him you mean whom I spared on the
Gorgonidonian plains, where Bumbomachides
Clytomestoridysarchides, the grandson of Neptune, was the chief
commander?
Artotrogus. I remember him; him, I suppose you mean, with the
golden armor, whose legions you puffed away with your breath, just
as the wind blows away leaves or the reed-thatched roof.
Pyrgopolinices. That, by my troth, was really nothing at all.
Artotrogus. Faith, that really was nothing at all in comparison with
other things I could mention (aside) which you never did. If any
person ever beheld a more perjured fellow than this, or one more full
in vain boasting, let him have me for himself: I’ll become his slave.
Pyrgopolinices. What are you saying?
Artotrogus. Why, that I remember in what fashion you broke the
foreleg of an elephant, in India, with your fist.
Pyrgopolinices. How—the foreleg?
Artotrogus. I meant to say the thigh.
Pyrgopolinices. I struck the blow without an effort.
Artotrogus. Troth, if, indeed, you had put forth your strength, your
arm would have passed right through the hide, the entrails, and the
frontispiece of the elephant.
Pyrgopolinices. I don’t care to talk about these things just now.
Artotrogus. I’ faith, ’tis really not worth while for you to tell me of it,
who know your prowess well. (Aside.) My appetite creates all these
tales. I must hear him right out with my ears, that my teeth mayn’t
have time to grow, and whatever lie he shall tell I must agree to it.
Pyrgopolinices. What was it I was saying?
Artotrogus. Oh, I know what you were going to say just now. I’
faith ’twas bravely done; I remember its being done.
Pyrgopolinices. What was that?
Artotrogus. Whatever it was you were going to say.
Pyrgopolinices. Have you got your tablets?
Artotrogus. Are you intending to enlist some one? I have them,
and a pen as well.
Pyrgopolinices. How quickly you guess my thoughts!
Artotrogus. ’Tis fit that I should study your inclinations, so that
whatever you wish should first occur to me.
Pyrgopolinices. What do you remember?
 Artotrogus. I do remember this: In Cilicia there were a hundred
and fifty men, a hundred in Cryphiolathronia, thirty at Sardis, sixty
men of Macedon, whom you slaughtered altogether in one day.
Pyrgopolinices. What is the sum total of those men?
Artotrogus. Seven thousand.
Pyrgopolinices. It must be as much; you keep the reckoning well.
Artotrogus. Yet I have none of them written down; still, I remember
it was so.
Pyrgopolinices. By my troth, you have a right good memory.
Artotrogus (aside). ’Tis the flesh-pots give it a fillip.
Pyrgopolinices. So long as you shall do as you have done
hitherto, you shall always have something to eat; I will always make
you a partaker at my table.
Artotrogus. Besides, in Cappadocia you would have killed five
hundred men altogether at one blow, had not your saber been blunt.
Pyrgopolinices. I let them live, because I was quite sick of fighting.
Artotrogus. Why should I tell you what all mortals know, that you,
Pyrgopolinices, live upon the earth with your valor, beauty, and
achievements unsurpassed? All the women are in love with you, and
that not without reason, since you are so handsome. Witness those
girls that pulled me by my mantle yesterday.
Pyrgopolinices. What was it they said to you?
Artotrogus. They questioned me about you. “Is Achilles here?”
says one to me. “No,” says I, “his brother is.” Then says the other to
me, “By my troth, but he is a handsome and a noble man. See how
his long hair becomes him! Certainly the women are lucky who share
his favors.”
Pyrgopolinices. And pray, did they really say so?
Artotrogus. They both entreated me to bring you past today, so
that they might see you.
 Pyrgopolinices. ’Tis really a very great plague to a man to be too
handsome!
Artotrogus. They are quite a nuisance to me; they are praying,
entreating, beseeching me to let them see you; sending for me for
that purpose, so that I can’t give my attention to your business.
Pyrgopolinices. It seems that it is time for us to go to the Forum,
that I may count out their pay to those soldiers whom I lately
enlisted; for King Seleucus entreated me with most earnest suit that I
would raise and enlist recruits for him. To that business I have
resolved to devote my attention this day.
Artotrogus. Come, let’s be going, then.
Pyrgopolinices. Guards, follow me.
—The Braggart Captain.

THE SUSPICIOUS MISER

megadorus and eunomia

Eunomia. Tell me pray, who is she whom you would like to take for
a wife?
Megadorus. I’ll tell you. Do you know that Euclio, the poor old man
close by?
Eunomia. I know him; not a bad sort of man.
Megadorus. I’d like his maiden daughter to be promised me in
marriage. Don’t make any words about it, sister; I know what you are
going to say—that she’s poor. This poor girl pleases me.
Eunomia. May the gods prosper it!
Megadorus. I hope the same.
Eunomia. Do you wish me to stay for anything else?
Megadorus. No; farewell.
Eunomia. And to you the same, brother.
(Goes into the house.)
 Megadorus. I’ll go to see Euclio, if he’s at home. But, ah! here
comes the very man toward his own house!
Enter Euclio

Euclio (to himself). I had a presentiment that I was going out to no

purpose when I left my house, and therefore I went unwillingly; for
neither did any one of the wardsmen come, nor yet the master of the
ward, who ought to have distributed the money. Now I’m making all
haste to hasten home; for, though I myself am here, my mind’s at
home.
Megadorus. May you be well, and ever fortunate, Euclio!
Euclio. May the gods bless you, Megadorus!
Megadorus. How are you? Are you quite well and contented?
Euclio (aside). It isn’t for nothing when a rich man accosts a poor
man courteously. Now, this fellow knows that I’ve got some gold; for
that reason he salutes me more courteously.
Megadorus. Do you say that you are well?
Euclio. Oh, I’m not very well in the money line.
Megadorus. But if you’ve a contented mind, you have enough for
passing a happy life with.
Euclio (aside). By my faith, the old woman has made a discovery
to him about the gold; it is clear she has told him. I’ll cut off her
tongue, and tear out her eyes, when I get home.
Megadorus. Why are you talking to yourself?
Euclio. I’m lamenting my poverty. I’ve a grown-up girl without a
portion, and one that can’t be disposed of in marriage; nor am I able
to marry her to anybody.
Megadorus. Hold your peace; be of good courage, Euclio; she
shall have a husband; you shall be assisted by myself. If you have
need of help, command me.
 Euclio (aside). Now he is aiming at my property, while he’s making
promises. He’s gaping for my gold, that he may devour it; in the one
hand he is carrying a stone, while he shows the bread in the other. I
trust no person who, rich himself, is exceedingly courteous to a poor
man; when he extends his hand with a kind air, then is he loading
you with some damage. I know these polyps, who, when they’ve
touched a thing, hold it fast.
Megadorus. Give me your attention, Euclio, for a little while; I wish
to speak a few words to you about a common concern of yours and
mine.
Euclio (aside). Alas! wo is me! My gold has been carried off from
my house. Now he’s wishing for this thing, I’m sure, to come to a
compromise with me; but I’ll look in my house first.
(He goes toward his door.)
Megadorus. Where are you going?
Euclio. I’ll return to you directly, for there’s something I must go
and see to at home.
(Goes into his house.)
Megadorus. I verily believe that when I make mention of his
daughter, for him to promise her to me, he’ll suppose that I am
laughing at him; for I do not know of any man poorer than he.
Euclio returns from his house

Euclio (aside). The gods favor me; my property’s all safe. If

nothing’s lost, it’s safe. I was dreadfully afraid before I went indoors. I
was almost dead. (Aloud.) I’m come back to you, Megadorus, if you
wish to say anything to me.
Megadorus. I thank you. I beg that as to what I shall inquire of
you, you’ll not hesitate to speak out boldly.
Euclio. So long, indeed, as you inquire nothing that I mayn’t
choose to speak out upon.
 Megadorus. Tell me, of what sort of family do you consider me to
be sprung?
Euclio. Of a good one.
Megadorus. What do you think about my character?
Euclio. It’s a good one.
Megadorus. What of my conduct?
Euclio. Neither bad nor dishonest.
Megadorus. Do you know my age?
Euclio. I know that you are as rich in years as in pocket.
Megadorus. I surely did always take you to be a citizen without
evil guile, and now I am convinced.
Euclio (aside). He smells the gold. (Aloud.) What do you want with
me now?
Megadorus. Since you know me, and I know you, what sort of
person you are, may it bring a blessing on myself, and you and your
daughter, if I now ask your daughter as my wife. Promise me that it
shall be so.
Euclio. Heyday! Megadorus, you are doing a deed that’s not
becoming to your usual actions, in laughing at me, a poor man, and
guiltless toward yourself and toward your family. For neither in act,
nor in words, have I ever deserved it of you that you should do what
you are doing now.
Megadorus. I vow that I neither came to laugh at you nor am I
laughing at you, nor do I think you deserving of it.
Euclio. Why, then, do you ask my daughter for yourself?
Megadorus. Because I believe that the match would be a good
thing for all of us.
Euclio. It suggests itself to my mind, Megadorus, that you are a
wealthy man, a man of rank, and that I am the poorest of the poor.
Now, if I should give my daughter in marriage to you, it suggests
itself to my mind that you are the ox, and that I am the ass; when I’m
yoked to you, and when I’m not able to bear the burden equally with
yourself, I, the ass, must lie down in the mire; you, the ox, would
regard me no more than if I had never been born. I should then feel
aggrieved, and my own class would laugh at me. In neither direction
should I have a fixed stall, if there should be a divorce; the asses
would tear me with their teeth, the oxen would butt at me with their
horns. This is the great risk, in my passing over from the asses to the
oxen.
Megadorus. The nearer you can unite yourself in alliance with
honorable people the better. Do you receive this proposal, listen to
me, and promise her to me.
Euclio. But there is no marriage portion, I tell you.
Megadorus. You are to give none; so long as she comes with
good principles, she is sufficiently portioned.
Euclio. I say so for this reason, that you mayn’t be supposing that
I have found any treasures.
Megadorus. I know that; don’t enlarge upon it. Promise her to me.
Euclio. So be it. (Starts and looks about.) But, oh, Jupiter, am I not
utterly undone?
Megadorus. What’s the matter with you?
Euclio. What was it sounded just now as though it were iron?
Megadorus. I ordered them to dig up the garden at my place.
(Euclio runs off into his house.) But where has this man gone? He’s
off, and he hasn’t fully answered me; he treats me with contempt.
Because he sees that I wish for his friendship, he acts after the usual
manner of mankind. For if a wealthy person goes to ask a favor of a
poorer one, the poor man is afraid to treat with him; through
suspicion he hurts his own interest. The same person, when this
opportunity is lost, afterward wishes for it too late.
Euclio (coming out of the house, addressing servant within). By
the powers, if I don’t give you up to have your tongue cut out by the
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!

ebookmass.com