Professional Documents
Culture Documents
PDF 14Th International Conference On Soft Computing Models in Industrial and Environmental Applications Soco 2019 Seville Spain May 13 15 2019 Proceedings Francisco Martinez Alvarez Ebook Full Chapter
PDF 14Th International Conference On Soft Computing Models in Industrial and Environmental Applications Soco 2019 Seville Spain May 13 15 2019 Proceedings Francisco Martinez Alvarez Ebook Full Chapter
https://textbookfull.com/product/biota-grow-2c-gather-2c-cook-
loucas/
https://textbookfull.com/product/10th-international-conference-
on-soft-computing-models-in-industrial-and-environmental-
applications-1st-edition-alvaro-herrero/
https://textbookfull.com/product/soft-computing-in-data-
science-5th-international-conference-scds-2019-iizuka-japan-
august-28-29-2019-proceedings-michael-w-berry/
https://textbookfull.com/product/business-information-
systems-22nd-international-conference-bis-2019-seville-spain-
june-26-28-2019-proceedings-part-ii-witold-abramowicz/
Business Information Systems: 22nd International
Conference, BIS 2019, Seville, Spain, June 26–28, 2019,
Proceedings, Part I Witold Abramowicz
https://textbookfull.com/product/business-information-
systems-22nd-international-conference-bis-2019-seville-spain-
june-26-28-2019-proceedings-part-i-witold-abramowicz/
https://textbookfull.com/product/proceedings-of-the-11th-
international-conference-on-soft-computing-and-pattern-
recognition-socpar-2019-ajith-abraham/
https://textbookfull.com/product/hybrid-artificial-intelligent-
systems-14th-international-conference-hais-2019-leon-spain-
september-4-6-2019-proceedings-hilde-perez-garcia/
https://textbookfull.com/product/proceedings-of-the-18th-online-
world-conference-on-soft-computing-in-industrial-applications-
wsc18-bernadetta-kwintiana-ane/
Advances in Intelligent Systems and Computing 950
Francisco Martínez Álvarez
Alicia Troncoso Lora
José António Sáez Muñoz
Héctor Quintián
Emilio Corchado Editors
14th International
Conference on Soft
Computing Models
in Industrial and
Environmental
Applications (SOCO 2019)
Seville, Spain, May 13–15, 2019,
Proceedings
Advances in Intelligent Systems and Computing
Volume 950
Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
Advisory Editors
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing,
Universidad Central de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, Electronic Engineering, University of Essex, Colchester, UK
László T. Kóczy, Department of Automation, Széchenyi István University,
Gyor, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas
at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao
Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information Technology,
University of Technology Sydney, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute
of Technology, Tijuana, Mexico
Nadia Nedjah, Department of Electronics Engineering, University of Rio de Janeiro,
Rio de Janeiro, Brazil
Ngoc Thanh Nguyen, Faculty of Computer Science and Management,
Wrocław University of Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering,
The Chinese University of Hong Kong, Shatin, Hong Kong
The series “Advances in Intelligent Systems and Computing” contains publications
on theory, applications, and design methods of Intelligent Systems and Intelligent
Computing. Virtually all disciplines such as engineering, natural sciences, computer
and information science, ICT, economics, business, e-commerce, environment,
healthcare, life science are covered. The list of topics spans all the areas of modern
intelligent systems and computing such as: computational intelligence, soft comput-
ing including neural networks, fuzzy systems, evolutionary computing and the fusion
of these paradigms, social intelligence, ambient intelligence, computational neuro-
science, artificial life, virtual worlds and society, cognitive science and systems,
Perception and Vision, DNA and immune based systems, self-organizing and
adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics
including human-machine teaming, knowledge-based paradigms, learning para-
digms, machine ethics, intelligent data analysis, knowledge management, intelligent
agents, intelligent decision making and support, intelligent network security, trust
management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are
primarily proceedings of important conferences, symposia and congresses. They
cover significant recent developments in the field, both of a foundational and
applicable character. An important characteristic feature of the series is the short
publication time and world-wide distribution. This permits a rapid and broad
dissemination of research results.
Editors
123
Editors
Francisco Martínez Álvarez Alicia Troncoso Lora
Data Science and Big Data Lab Data Science and Big Data Lab
Pablo de Olavide University Pablo de Olavide University
Seville, Spain Seville, Spain
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
v
vi Preface
Particular thanks go as well to the conference main sponsors, Startup Ole and
IEEE SMC Spanish Chapter, who jointly contributed in an active and constructive
manner to the success of this initiative.
We would like to thank all the special session organizers, contributing authors,
as well as the members of the Program Committees and the Local Organizing
Committee for their hard and highly valuable work. Their work has helped to
contribute to the success of the SOCO 2019 event.
General Chairs
vii
viii Organization
Program Committee
Special Sessions
Soft Computing Methods in Manufacturing and Management Systems
Program Committee
Program Committee
Program Committee
Program Committee
Machine Learning
Indexes to Find the Optimal Number of Clusters
in a Hierarchical Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
José David Martín-Fernández, José María Luna-Romera, Beatriz Pontes,
and José C. Riquelme-Santos
Analysis and Application of Normalization Methods
with Supervised Feature Weighting to Improve K-means Accuracy . . . . 14
Iratxe Niño-Adan, Itziar Landa-Torres, Eva Portillo,
and Diana Manjarres
Classifying Excavator Operations with Fusion Network
of Multi-modal Deep Learning Models . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Jin-Young Kim and Sung-Bae Cho
A Study on Trust in Black Box Models and Post-hoc Explanations . . . . 35
Nadia El Bekri, Jasmin Kling, and Marco F. Huber
A Study on Hyperparameter Configuration for Human
Activity Recognition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Kemilly D. Garcia, Tiago Carvalho, João Mendes-Moreira,
João M. P. Cardoso, and André C. P. L. F. de Carvalho
A Fuzzy Approach for Sentences Relevance Assessment
in Multi-document Summarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Eduardo Valladares-Valdés, Alfredo Simón-Cuevas, José A. Olivas,
and Francisco P. Romero
Online Estimation of the State of Health of a Rechargeable Battery
Through Distal Learning of a Fuzzy Model . . . . . . . . . . . . . . . . . . . . . . 68
Luciano Sánchez, José Otero, Manuela González, David Anseán,
and Inés Couso
xv
xvi Contents
1 Introduction
In recent years, the size of the information available for various types of studies
has grown considerably. Areas like medicine [1], social networks [2], energy [3]
or electronic consumption [4] are just a few examples of this, with an increasing
amount of data. This information needs to be processed to get some useful
knowledge.
Among the different possible solutions to data analysis we focus on Machine
Learning techniques, allowing us to extract the main features and a model cov-
ering the main information in a dataset. One of the most used model is called
clustering, which determines the number of instances of a certain grouping within
the data under study. Within the existing grouping variants, hierarchical clus-
tering provides us with very interesting additional information. We can see the
c Springer Nature Switzerland AG 2020
F. Martı́nez Álvarez et al. (Eds.): SOCO 2019, AISC 950, pp. 3–13, 2020.
https://doi.org/10.1007/978-3-030-20055-8_1
4 J. D. Martı́n-Fernández et al.
evolution of the clusters in each step of the algorithm, thus studying the group-
ing of X elements within the data. There exists two subtypes within hierarchical
clustering: dissociative or descending, starting from a group with all the ele-
ments, and ending in a cluster for each instance in the dataset; or agglomerative
or ascending, starting with as many clusters as exists elements in the dataset
and ending with a single agglomerative cluster with all of them.
Nowadays, there exists several frameworks to work with Machine Learning
techniques to obtain knowledge. One of the most known is Apache Hadoop [5],
that is built around the programming model based on the Google paradigm
MapReduce [6]. Moreover, one of the most widely used open source projects
is Apache Spark [7]. In the Google paradigm, it is read and written from the
hard disk on many occasions, which reduces produces a detriment in the speed
of data processing. Spark, the number of write/read cycles on the disk, so that
intermediate calculations are logically and quickly stored in RAM. To do this,
Spark uses a data structure called “Resilient Distributed Datasets” (RDD), that
are specially designed to parallelize cache calculations with high data volume. In
addition, this system contains the scalable library for Machine Learning (MLlib),
with a series of such as algorithms classification, regression, recommendation
systems and clustering techniques, will be of great help to achieve our goal [8].
The purpose of this article is to present a new agglomerative clustering tech-
nique implemented in Apache Spark. We have tested our algorithm using diverse
datasets, that were created by means of a random database generator. Further-
more, we have applied different internal clustering validation indexes (CVIs)
[9] in order to test our clustering results and compare the CVIs performance
between the agglomerative hierarchical clustering implemented (AHC) and that
provided by Spark as a dissociative modality, Bisecting K-Means (BK-Means)
[10].
The rest of the document is organized as follows. Section 2 reviews different
types of clustering techniques, as well as the CVIs used during our experimenta-
tion. Section 3 describes the algorithm and its implementation. Section 4 presents
the experiments carried out and, finally, Sect. 5 summarizes the main conclusions
of this work.
2 Related Work
In this section are reviewed the main grouping methods, as well as the internal
validation indexes that have been used in our experimentation.
There are several types of clustering algorithms, which could be classified into
the following categories depending on the method we use [11]:
Within the family of hierarchical algorithms there are two versions or strate-
gies that can be used:
From the first group of algorithms, we can find numerous examples in the
literature [12–14]. However, in this work we present a version of the agglomerative
option. The main problem of this implementation lies in the computation time
needed when treating with large amount of data. Hence, there are few examples
in the literature of implementations of this type of strategies [15,16].
Dunn and Silhouette. We describe in the following the two most used vali-
dation indexes in the literature, which have been implemented in this work for
our experiments:
6 J. D. Martı́n-Fernández et al.
3 Our Proposal
In this section we present our approach for AHC. Our technique starts from as
many clusters as instances and, in an ascending way, groups them until it reaches
a single cluster. In the next, we show the pseudocode of our strategy is shown
in Algorithm 1:
The algorithm receives as parameters: a RDD of objects of type “Distance”
[20], (class created internally to represent the distance between any two ele-
ments of the database); the number of clusters to be obtained; the strategy for
computing the distances; and the total number of instances in the dataset.
Line 10 refers to the calculation of the Cartesian product, which is neces-
sary to find the distances between the points or clusters of each iteration with
respect to the other elements of the RDD of objects of type “Distance”. In addi-
tion, the step performed on line 11 is configurable according to the designated
strategy for calculating the distance between elements in the database, being
Indexes Hierarchical Clustering 7
the implemented options “min”, “max” and “avg”. They refer to the minimum,
maximum or mean distance between the distances of the remaining points from
each of the points that make up the pair found during line 2 of the algorithm,
respectively. In the literature, each of these strategies establishes a different hier-
archical clustering typology. Being “minimum or simple link grouping” (simple
linkage), “maximum or complete link grouping” (complete linkage) and “average
or average link grouping” (average linkage), respectively [21].
3.1 Implementation
For the creation of this hierarchical grouping, several variants can be made
depending on the basis for storing the information, using “Resilient Distributed
Dataset” (RDD) or “DataFrames”. Both objects are provided by Apache Spark.
In addition, some functions from MLlib were used, which allows us to delegate
some calculations of our algorithm.
Following Spark recommendations, collect() and coalesce() [22] methods were
used in order to accelerate the process. With the collect() method, it is possible
to obtain data stored in memory during previous calculations, so that it is not
necessary to wait for the executions lazy of the Spark framework. Through the
8 J. D. Martı́n-Fernández et al.
4 Experimentation
In this section we present the experimental setup and results obtained using our
AHC approach and the comparison with respect to the use of the dissociative
clustering algorithm BK-Means.
Our goal is to check the goodness of the different grouping obtained by our algo-
rithm, using several datasets and evaluating the results by means of multiple
CVIs. The experiments were executed in: IntelliJ IDEA development environ-
ment; the Apache Spark framework using the Scala language; and the Machine
Learning library provided by the MLlib framework; a computer with an Intel
Core i7-7700HQ CPU with 4 cores of 2.8 GHz, 16 GB of RAM, an SSD of 256 GB
and a HDD of 1 TB.
A total of 60 datasets have been used in our experimentation, which were
generated by using the database generator in [9]. This tool allowed us to configure
the desired number of clusters, dimensions, and the number of points for each
cluster.
For the experimentation, three different configurations for the number of
clusters (K) have been used: 3, 5 and 7; 20 different configurations for the dimen-
sionality of the data: from 1 to 20; and 100 points for each of cluster. In order to
achieve the 20 different dimensions expressed above, a dataset has been taken as
the basis for each different K with 20 total dimensions, from which the different
dimensions from 1 to 20 have been selected.
Indexes Hierarchical Clustering 9
In order to study which CVI offers the best results using our hierarchical algo-
rithm as the basis for clustering, we must first define how to measure this good-
ness. In our experiments, the modality “avg” distance of the hierarchical clus-
tering algorithm has been chosen, explained in Sect. 3. As for the configuration
of the BK-Means algorithm, all the default values for parameters have been set.
Since the number of clusters (K) for each of the datasets is known beforehand,
we would check in how many datasets each of the indexes matches K. After
executing the algorithm for each of the datasets, different cluster numbers on
the resulting model for each CVI has been tested. Specifically, K values from
3 to 9 have been used (3 < K < 9 ). This interval has been chosen in order to
guarantee that all K values can be found, since, the minimum would be 3 and
the maximum would be 7 in the data used during the experimentation.
The results have been grouped according to two criteria: by the number of
clusters, and by the number of dimensions in each of the databases. Following the
first of the criteria, each index could obtain a maximum of 20 hits in each of the
numbers of clusters. Whereas for the second criterion, each index could obtain a
maximum of 3 hits for each of the dimensions available in our databases. In this
sense, each of the indexes will have two different hits: one grouping all the cluster
numbers for each dimension, being able to obtain a maximum of 20; and another
grouping all dimensions for each cluster number, being able to obtain a maximum
of 3. The results can be summarized in the following Tables 1, 2 and 3:
Table 1. Summary of hits of each index grouped by the number of clusters in each
dataset.
Table 2. Summary of hits of each index grouped by the number of dimension in each
dataset in AHC execution.
Index 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Total
Silhouette 1 1 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 14
Dunn 1 1 1 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 14
Silhouette-BD 0 0 1 2 1 2 3 2 3 2 3 3 2 2 2 2 2 3 3 3 7
Dunn-BD 0 1 2 2 2 2 3 3 2 3 3 3 3 3 3 3 3 3 3 3 13
Davis-Bouldin 1 1 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 14
Another random document with
no related content on Scribd:
condition of things than I knew, and had a deeper hold upon public
attention than I had supposed. Invitations began to pour in upon me
from colleges, lyceums, and literary societies, offering me one
hundred, and even two hundred dollars for a single lecture.
I had, sometime before, prepared a lecture on “Self-made men,”
and also one upon Ethnology, with special reference to Africa. The
latter had cost me much labor, though as I now look back upon it, it
was a very defective production. I wrote it at the instance of my
friend Doctor M. B. Anderson, President of Rochester University,
himself a distinguished Ethnologist, a deep thinker and scholar. I had
been invited by one of the literary societies of Western Reserve
College (then at Hudson, but recently removed to Cleveland, Ohio),
to address it on Commencement day; and never having spoken on
such an occasion, never, indeed, having been myself inside of a
school-house for the purpose of an education, I hesitated about
accepting the invitation, and finally called upon Prof. Henry Wayland,
son of the great Doctor Wayland of Brown University, and on Doctor
Anderson, and asked their advice whether I ought to accept. Both
gentlemen advised me to do so. They knew me, and evidently
thought well of my ability. But the puzzling question now was, what
shall I say if I do go there? It won’t do to give them an old-fashioned
anti-slavery discourse. (I learned afterwards that such a discourse
was precisely what they needed, though not what they wished; for
the faculty, including the President, was in great distress because I,
a colored man, had been invited, and because of the reproach this
circumstance might bring upon the College.) But what shall I talk
about? became the difficult question. I finally hit upon the one before
mentioned. I had read, when in England a few years before, with
great interest, parts of Doctor Pritchard’s “Natural History of Man,” a
large volume marvelously calm and philosophical in its discussion of
the science of the origin of the races, and was thus in the line of my
then convictions. I sought this valuable book at once in our
bookstores, but could not obtain it anywhere in this country. I sent to
England, where I paid the sum of seven and a half dollars for it. In
addition to this valuable work, President Anderson kindly gave me a
little book entitled, “Man and His Migrations,” by Dr. R. G. Latham,
and loaned me the large work of Dr. Morton the famous
Archaeologist, and that of Messrs. Nott and Glidden, the latter
written evidently to degrade the negro and support the then
prevalent Calhoun doctrine of the rightfulness of slavery. With these
books, and occasional suggestions from Dr. Anderson and Prof.
Wayland, I set about preparing my Commencement address. For
many days and nights I toiled, and succeeded at last in getting
something together in due form. Written orations had not been in my
line. I had usually depended upon my unsystematized knowledge,
and the inspiration of the hour and the occasion; but I had now got
the “scholar bee in my bonnet,” and supposed that inasmuch as I
was to speak to college professors and students, I must at least
make a show of some familiarity with letters. It proved, as to its
immediate effect, a great mistake, for my carefully studied and
written address, full of learned quotations, fell dead at my feet, while
a few remarks I made extemporaneously at collation, were
enthusiastically received. Nevertheless, the reading and labor
expended were of much value to me. They were needed steps
preparatory to the work upon which I was about to enter. If they
failed at the beginning, they helped to success in the end. My lecture
on “The Races of Men” was seldom called for, but that on “Self-made
Men” was in great demand, especially through the West. I found that
the success of a lecturer depends more upon the quality of his stock
in store, than the amount. My friend, Wendell Phillips (for such I
esteem him), who has said more cheering words to me, and in
vindication of my race, than any man now living, has delivered his
famous lecture on the “Lost Arts” during the last forty years; and I
doubt if among all his lectures, and he has many, there is one in
such requisition as this. When Daniel O’Connell was asked why he
did not make a new speech he playfully replied, that “it would take
Ireland twenty years to learn his old ones.” Upon some such
consideration as this, I adhered pretty closely to my old lecture on
“Self-made Men,” retouching and shading it a little from time to time
as occasion seemed to require.
Here, then, was a new vocation before me, full of advantages,
mentally and pecuniarily. When in the employment of the American
Anti-Slavery Society, my salary was about four hundred and fifty
dollars a year, and I felt I was well paid for my services; but I could
now make from fifty to a hundred dollars a night, and have the
satisfaction, too, that I was in some small measure helping to lift my
race into consideration; for no man who lives at all, lives unto
himself; he either helps or hinders all who are in anywise connected
with him. I never rise to speak before an American audience without
something of the feeling that my failure or success will bring blame
or benefit to my whole race. But my activities were not now confined
entirely to lectures before lyceums. Though slavery was abolished,
the wrongs of my people were not ended. Though they were not
slaves they were not yet quite free. No man can be truly free whose
liberty is dependent upon the thought, feeling, and action of others;
and who has himself no means in his own hands for guarding,
protecting, defending, and maintaining that liberty. Yet the negro after
his emancipation was precisely in this state of destitution. The law on
the side of freedom is of great advantage only where there is power
to make that law respected. I know no class of my fellowmen,
however just, enlightened, and humane, which can be wisely and
safely trusted absolutely with the liberties of any other class.
Protestants are excellent people, but it would not be wise for
Catholics to depend entirely upon them to look after their rights and
interests. Catholics are a pretty good sort of people (though there is
a soul-shuddering history behind them), yet no enlightened
Protestants would commit their liberty to their care and keeping. And
yet the government had left the freedmen in a worse condition than
either of these. It felt that it had done enough for him. It had made
him free, and henceforth he must make his own way in the world, or
as the slang phrase has it, “Root, pig, or die”; yet he had none of the
conditions for self-preservation or self-protection. He was free from
the individual master, but the slave of society. He had neither
property, money, nor friends. He was free from the old plantation, but
he had nothing but the dusty road under his feet. He was free from
the old quarter that once gave him shelter, but a slave to the rains of
summer and the frosts of winter. He was in a word literally turned
loose naked, hungry, and destitute to the open sky. The first feeling
towards him by the old master classes, was full of bitterness and
wrath. They resented his emancipation as an act of hostility towards
them, and since they could not punish the emancipator, they felt like
punishing the object which that act had emancipated. Hence they
drove him off the old plantation, and told him he was no longer
wanted there. They not only hated him because he had been freed
as a punishment to them, but because they felt that they had been
robbed of his labor. An element of greater bitterness still came into
their hearts: the freedman had been the friend of the Government,
and many of his class had borne arms against them during the war.
The thought of paying cash for labor that they could formerly extort
by the lash did not in anywise improve their disposition to the
emancipated slave, or improve his own condition. Now, since poverty
has, and can have no chance against wealth, the landless against
the land owner, the ignorant against the intelligent, the freedman was
powerless. He had nothing left him but a slavery-distorted and
diseased body, and lame and twisted limbs with which to fight the
battle of life. I, therefore, soon found that the negro had still a cause,
and that he needed my voice and pen with others to plead for it. The
American Anti-Slavery Society, under the lead of Mr. Garrison, had
disbanded, its newspapers were discontinued, its agents were
withdrawn from the field, and all systematic efforts by abolitionists
were abandoned. Many of the Society, Mr. Phillips and myself
amongst the number, differed from Mr. Garrison as to the wisdom of
this course. I felt that the work of the Society was not done, that it
had not fulfilled its mission, which was not merely to emancipate, but
to elevate the enslaved class; but against Mr. Garrison’s leadership
and the surprise and joy occasioned by the emancipation, it was
impossible to keep the association alive, and the cause of the
freedmen was left mainly to individual effort and to hastily
extemporized societies of an ephemeral character, brought together
under benevolent impulse, but having no history behind them, and
being new to the work, they were not as effective for good as the old
society would have been had it followed up its work and kept its old
instrumentalities in operation.
From the first I saw no chance of bettering the condition of the
freedman, until he should cease to be merely a freedman, and
should become a citizen. I insisted that there was no safety for him,
or for any body else in America, outside the American Government:
that to guard, protect, and maintain his liberty, the freedman should
have the ballot; that the liberties of the American people were
dependent upon the Ballot-box, the Jury-box, and the Cartridge-box,
that without these no class of people could live and flourish in this
country, and this was now the word for the hour with me, and the
word to which the people of the north willingly listened when I spoke.
Hence regarding as I did, the elective franchise as the one great
power by which all civil rights are obtained, enjoyed, and maintained
under our form of government, and the one without which freedom to
any class is delusive if not impossible, I set myself to work with
whatever force and energy I possessed to secure this power for the
recently emancipated millions.
Wendell Phillips
The demand for the ballot was such a vast advance upon the
former objects proclaimed by the friends of the colored race, that it
startled and struck men as preposterous and wholly inadmissible.
Anti-slavery men themselves were not united as to the wisdom of
such demand. Mr. Garrison himself, though foremost for the abolition
of slavery, was not yet quite ready to join this advanced movement.
In this respect he was in the rear of Mr. Phillips; who saw not only
the justice, but the wisdom and necessity of the measure. To his
credit it may be said, that he gave the full strength of his character
and eloquence to its adoption. While Mr. Garrison thought it too
much to ask, Mr. Phillips thought it too little. While the one thought it
might be postponed to the future, the other thought it ought to be
done at once. But Mr. Garrison was not a man to lag far in the rear of
truth and right, and he soon came to see with the rest of us that the
ballot was essential to the freedom of the freedman. A man’s head
will not long remain wrong, when his heart is right. The applause
awarded to Mr. Garrison by the conservatives, for his moderation
both in respect of his views on this question, and the disbandment of
the American Anti-Slavery Society must have disturbed him. He was
at any rate soon found on the right side of the suffrage question.
The enfranchisement of the freedmen was resisted on many
grounds, but mainly these two: first the tendency of the measure to
bring the freedmen into conflict with the old master-class, and the
white people of the South generally. Secondly, their unfitness, by
reason of their ignorance, servility, and degradation, to exercise so
great a power as the ballot, over the destinies of this great nation.
These reasons against the measure which were supposed to be
unanswerable, were in some sense the most powerful arguments in
its favor. The argument that the possession of suffrage would be
likely to bring the negro into conflict with the old master-class at the
South, had its main force in the admission that the interests of the
two classes antagonized each other and that the maintenance of the
one would prove inimical to the other. It resolved itself into this, if the
negro had the means of protecting his civil rights, those who had
formerly denied him these rights would be offended and make war
upon him. Experience has shown in a measure the correctness of
this position. The old master was offended to find the negro whom
he lately possessed the right to enslave and flog to toil, casting a
ballot equal to his own, and resorted to all sorts of meanness,
violence, and crime, to dispossess him of the enjoyment of this point
of equality. In this respect the exercise of the right of suffrage by the
negro has been attended with the evil, which the opponents of the
measure predicted, and they could say “I’ve told you so,” but
immeasurably and intolerably greater would have been the evil
consequences resulting from the denial to one class of this natural
means of protection, and granting it to the other, and hostile class. It
would have been, to have committed the lamb to the care of the wolf
—the arming of one class and disarming the other—protecting one
interest, and destroying the other—making the rich strong, and the
poor weak—the white man a tyrant, and the black man a slave. The
very fact therefore that the old master-classes of the South felt that
their interests were opposed to those of the freedmen, instead of
being a reason against their enfranchisement, was the most powerful
one in its favor. Until it shall be safe to leave the lamb in the hold of
the lion, the laborer in the power of the capitalist, the poor in the
hands of the rich, it will not be safe to leave a newly emancipated
people completely in the power of their former masters, especially
when such masters have not ceased to be such from enlightened
moral convictions but by irresistible force. Then on the part of the
Government itself, had it denied this great right to the freedmen, it
would have been another proof that “Republics are ungrateful”. It
would have been rewarding its enemies, and punishing its friends—
embracing its foes, and spurning its allies,—setting a premium on
treason, and degrading loyalty. As to the second point, viz.: the
negro’s ignorance and degradation, there was no disputing either. It
was the nature of slavery from whose depths he had arisen to make
him so, and it would have kept it so. It was the policy of the system
to keep him both ignorant and degraded, the better and more safely
to defraud him of his hard earnings; and this argument never
staggered me. The ballot in the hands of the negro was necessary to
open the door of the school house, and to unlock the treasures of
knowledge to him. Granting all that was said of his ignorance, I used
to say, “if the negro knows enough to fight for his country he knows
enough to vote; if he knows enough to pay taxes for the support of
the government, he knows enough to vote; if he knows as much
when sober, as an Irishman knows when drunk, he knows enough to
vote.”
And now while I am not blind to the evils which have thus far
attended the enfranchisement of the colored people, I hold that the
evils from which we escaped, and the good we have derived from
that act, amply vindicate its wisdom. The evils it brought are in their
nature temporary, and the good is permanent. The one is
comparatively small, the other absolutely great. The young child has
staggered on his little legs, and he has sometimes fallen and hurt his
head in the fall, but then he has learned to walk. The boy in the
water came near drowning, but then he has learned to swim. Great
changes in the relations of mankind can never come, without evils
analogous to those which have attended the emancipation and
enfranchisement of the colored people of the United States. I am
less amazed at these evils, than by the rapidity with which they are
subsiding and not more astonished at the facility with which the
former slave has become a free man, than at the rapid adjustment of
the master-class to the new situation.
Unlike the movement for the abolition of Slavery, the success of
the effort for the enfranchisement of the freedmen was not long
delayed. It is another illustration of how any advance in pursuance of
a right principle, prepares and makes easy the way to another. The
way of transgression is a bottomless pit, one step in that direction
invites the next, and the end is never reached; and it is the same
with the path of righteous obedience. Two hundred years ago, the
pious Doctor Godwin dared affirm that it was “not a sin to baptize a
negro,” and won for him the rite of baptism. It was a small
concession to his manhood; but it was strongly resisted by the
slaveholders of Jamaica, and Virginia. In this they were logical in
their argument, but they were not logical in their object. They saw
plainly that to concede the negro’s right to baptism was to receive
him into the Christian Church, and make him a brother in Christ; and
hence they opposed the first step sternly and bitterly. So long as they
could keep him beyond the circle of human brotherhood, they could
scourge him to toil, as a beast of burden, with a good Christian
conscience, and without reproach. “What!” said they, “baptize a
negro? preposterous!” Nevertheless the negro was baptized and
admitted to church fellowship; and though for a long time his soul
belonged to God, his body to his master, and he poor fellow had
nothing left for himself, he is at last not only baptized, but
emancipated and enfranchised.
In this achievement, an interview with President Andrew
Johnson, on the 7th of February, 1866, by a delegation consisting of
George T. Downing, Lewis H. Douglass, Wm. E. Matthews, John
Jones, John F. Cook, Joseph E. Otis, A. W. Ross, William Whipper,
John M. Brown, Alexander Dunlop, and myself, will take its place in
history as one of the first steps. What was said on that occasion
brought the whole question virtually before the American people.
Until that interview the country was not fully aware of the intentions
and policy of President Johnson on the subject of reconstruction,
especially in respect of the newly emancipated class of the South.
After having heard the brief addresses made to him by Mr. Downing
and myself, he occupied at least three quarters of an hour in what
seemed a set speech, and refused to listen to any reply on our part,
although solicited to grant a few moments for that purpose. Seeing
the advantage that Mr. Johnson would have over us in getting his
speech paraded before the country in the morning papers, the
members of the delegation met on the evening of that day, and
instructed me to prepare a brief reply which should go out to the
country simultaneously with the President’s speech to us. Since this
reply indicates the points of difference between the President and
ourselves, I produce it here as a part of the history of the times, it
being concurred in by all the members of the delegation.
Both the speech and the reply were commented upon very
extensively.
From this time onward, the question of suffrage for the freedmen,
was not allowed to rest. The rapidity with which it gained strength,
was something quite marvelous and surprising even to its advocates.
Senator Charles Sumner soon took up the subject in the Senate and
treated it in his usually able and exhaustive manner. It was a great
treat to listen to his argument running through two days, abounding
as it did in eloquence, learning, and conclusive reasoning. A
committee of the Senate had reported a proposition giving to the
States lately in rebellion in so many words complete option as to the
enfranchisement of their colored citizens: only coupling with that
proposition the condition that, to such States as chose to enfranchise
such citizens, the basis of their representation in Congress should be
proportionately increased; or, in other words, only three-fifths of the
colored citizens should be counted in the basis of representation in
States where colored citizens were not allowed to vote, while in the
States granting suffrage to colored citizens, the entire colored people
should be counted in the basis of representation. Against this
proposition, myself and associates addressed to the Senate of the
United States the following memorial: