The following is the written testimony submitted by computer expert Robert Strunk to Judge Richard Niehaus after the Judge had ruled in 1985 that the Cincinnatus PAC's suit against the Hamilton County Board of Elections was correct.
The Judge had stated in his ruling that: "There is no adequate and proper safeguard to prevent the computers from being programmed to distort the election results." This is one of the most superb statements ever issued by any computer expert since computers began to be used to count votes in the United States circa 1974.
The system Strunk is warning against is the CES IBM punch card system which is still being used in much of the nation in 1996. Despite Strunk's devastating expose of the weaknesses of the system, the Judge failed to provide a remedy to the wrong he had diagnosed, and the system is still used in Cincinnati as of this writing in May of 1996. - Jim Condit Jr.
COURT OF COMMON
HAMILTON COUNTY, OHIO
JERRY SCHUTZMAN, Plaintiff
GEORGE C. EYRICH, EARL
BARNES, SIDNEY WEIL AND JOHN A. WIETHE,
AS MEMBERS OF THE BOARD OF ELECTIONS, HAMILTON COUNTY, OHIO and
THE CITY OF CINCINNATI, Defendants
Case No. A8108370
STATE OF OHIO
COUNTY OF HAMILTON
AFFIDAVID OF ROBERT C. STRUNK AS EXPERT WITNESS FOR PLAINTIFF
I am Robert C. Strunk, an adult resident of [omitted from internet] Kentucky, and being first duly sworn say that this Affidavit is given voluntarily on my past and I am not related by blood or marriage to any of the parties to this litigation and first came to know the Plaintiff and his counsel in connection with this action in October, 1985. Further, I give this Affidavit to assist this Court with the computer science and related technical aspects which are before the Court in this case.
Further, I state in this Affidavit:
1. That nothing in this Affidavit is intended, nor should be construed, to attack or question the integrity of any particular Defendant or to accuse any particular person of any criminal activity or other wrongdoing. In fact, I feel that those persons I observed in the course of my investigation all seemed to be sincerely interested in doing their jobs conscientiously.
2. That I have been engaged by the Plaintiff as an expert in computer-related matters including, especially, the programming of computers. My Curriculum Vitae and credentials are attached hereto as Exhibit A. Briefly, for the past 20 years, and thus for my entire occupational career, I have been directly and entirely involved in the fields of Computer Science and Data Processing.
3. That, pursuant to this Court's order certified on November 4, 1985, I have had access to all phases of the procedure used by the Hamilton County Board of Elections (hereinafter referred to as the "BOE") to arrive at and certify the vote count in the recent general election of November 5, 1985 (hereinafter referred to as the "election"). My observations began with the preliminary test of the computer vote counting process on October 31, 1985. I personally observed the voting process at a specific precinct, at the BOE office on Sycamore Street, and at the Regional Computer Center (hereinafter referred to as the "RCC") several blocks away in the Temple Bar Building at Court and Main Streets. I also personally observed the procedures for transmitting the ballots and other election materials necessary for the vote count among these various locations.
4. That the opinions I am expressing are based solely on these personal observations and the inquiries which I made at the time of various staff persons employed at the sites. In no way are any of these opinions based on the opinions, observations, or representations of any other expert witnesses or others involved in this case.
5. That I have been provided with a copy of this Court's "Findings of Fact and Conclusions of Law". Some of these findings fall within my area of competence and expertise and I, therefore, initially observe that my findings are consistent with the Court's in the following areas:
a. That the entity which actually carried out the computerized vote count in Hamilton County is the RCC. The RCC is both the lessee of the computer equipment and the employer of all of the people involved in the actual counting of the ballots cast in the election.
b. That the RCC was not to my observation supervised in the election either by the BOE or by the office of the Ohio Secretary of State. Although the BOE had two witnesses at the RCC the night o the election, they could undoubtedly only testify to the fact that the ballot tapes were received from the appointed messengers. In my opinion, these witnesses would have to exercise their faith in order to be able to testify that the ballot tapes they received were actually used in the vote counting process at the RCC. Do not construe this remark to mean that these witnesses were foolish people - that would be untrue. The idea that I am trying to convey is that it is probably impossible for an observer to tell what is going on inside of a computer system; it would take a highly-trained technician quite a length of time to certify that there was no interference with the proposed processes. For example, it is very easy to make a computer program seemingly read the data on a reel of tape, then completely ignore that data and print out some pre-arranged answers. The BOE witnesses could not possibly have detected any vote fraud at this level. By way of explanation, I mean a highly-trained technician on both the hardware and the software aspects of the operating system actually used at the RCC vote counting process. Even having this expertise the technician would have to be there and examine the hardware and observe all of the actions of the operators continuously.
c. That Cincinnati City Council candidates' votes were among those votes which were counted by the RCC in this election.
d. That no one on either the BOE or RCC staff was involved in the planning or writing of the computer program that actually counted the ballots in the election. The program was provided, I am told, by an unaffiliated entity known as Computer Election Systems (hereinafter referred to as "CES") headquartered in San Francisco, California. I examined a printed copy of the Cobol Source statements for this program and found it to be written in a style that is very difficult to understand. I do not claim to have an understanding of this program from my reading of it. In fact, I found no one who was willing to say that he understood the internal operations of this program. While I hesitate to say that it is essential to have someone understand how the program was written in order to make use of the program, I can say that it would probably be difficult to find any hidden traps that a clever programmer might plant in a program as confusing as this one. Good programming today calls for "Structured Programming" techniques that make programs relatively easy to read and understand. This program is not a structured program.
e. That, because the RCC computer system has many terminals attached to it, the vote counting system is open to alteration by parties unseen and unknown to any observers. These terminals are both local (usually within the same building) and remote (accessed by modem over both private lines and the switched network). The staff at the RCC seemed sensitive to this possibility and seemed to take the necessary steps to prevent any type of fraud by this means. The point, however, is that it would require an extraordinary type of observer from the BOE to detect system access or alteration through any terminal that he could not actually see.
f. That there is at least one staff position at the RCC which would afford someone a definite opportunity to alter the computer program that counts the ballots and, therefore, to alter the results of the election itself. There is no one on the staff of the BOE who is charged with the care of this computer program nor is there any such person on its staff who is able to care for this program. It requires no stretch of the imagination to say that this program has but one keeper locally and that he reports only to the RCC.
g. That the BOE is directly involved only in the part of this computerized system that initially reads the ballot cards onto the computer tapes. This portion of the process is a rather simple part of the process. However, I would like to point out that there is another computer program involved in this process. While I take it that this is a relatively uncomplicated program, it seems worthwhile to point out that no one seems to even have a printed copy of this program. This program was provided by CES, the same company that provided the program that counts the votes at the RCC. Using each of these programs to validate the vote count of the other is a little like sending the fox out to watch the chickens.
6. That it would require an act of faith for the members of the BOE to "certify" that the vote count generated by this computerized system is a true and accurate county. While I cannot comment on the legality of what I have observed, it does seem to me that the BOE has delegated its responsibility over the computerized vote counting process to (a single staff member of) the RCC. This delegation may seem entirely appropriate to some because similar decisions are made every day in the routine use of millions of computers across the country. There is one tremendous difference, however, in this particular instance. That difference could be characterized by the word "accountability". Computers (and their programmers) that perform accounting tasks are held accountable by the principles of accounting. Computers that keep track of inventories are held accountable by a rotating scheme of actual physical counting performed by real human beings. Computers that do word processing are held accountable by the people striking the keys on the keyboards. Computers that keep track of votes cast in an election are held accountable by voters, it's true, but since voting must be kept secret to be successful, the voters must turn over their rights to the voting instrument itself. In other words, computers that keep track of votes cast in an election are held accountable only by the punched cards that they are to count. It seems clear that the definition of "accountability" suffers in this instance.
In summary, therefore, I conclude my opinions with the following observations:
1. There is no understanding of the program governing the computer operations at the BOE on Sycamore Street, by any of the BOE staff. The BOE staff is therefore utilizing a "black box" at its site. By "black box" I mean the BOE staff is relegated to putting information in the "black box" (computer) and later accepting data from this "black box" (computer) with no understanding or control over what goes on inside the box (computer). The BOE is therefore out of understanding and ability to supervise th computer phase which occurs at its own site on Sycamore Street, at the hands of its own staff. The BOE is completely dependent, therefore, on others, in this case RCC and CES, and, therefore is operating on what I will call "blind faith".
2. Similarly, moving one step out of the BOE to the RCC staff, the RCC staff is accepting the CES program and is also exercising "blind faith" in a program prepared by another entity (CES). Therefore, again, this produces the "black box" reality also at the RCC, i.e. RCC staff is relegated to putting information in the "black box" (computer) and accepting data from the "black box" (computer) with no understanding or control over what goes on inside the "black box" (computer).
3. It follows therefore that the "certification" by the BOE of the accuracy of the election vote count (BOE certifications as to this election are attached hereto as Exhibit B), is even more an exercise "in blind faith". That is to say that the certifications contained in Exhibit B are based upon the Board's own act of faith as described above in 1, and, upon the RCC's act of faith described in 2 above. I am therefore able to support this Court's finding number 7 that
"There is no adequate and proper safeguard against the computers being programmed to distort the election results."
4. It is necessary to reemphasize that my opinions here sound in the vulnerability of the existing system and do not allege wrongdoing. Indeed, vote count manipulation has occurred, the probability of it not being detected in the present system is very high.
5. I therefore also conclude that the existing system, as it is in place in Hamilton County, is not verifiable absent a complete hand count. Therefore, we all indulge in "blind faith" when we defer to the present computerized vote count system relied upon by the BOE. Nor does it encourage me to state that there is actually no way that I or any other individual can guarantee the accuracy of a computer vote count unless I have personally and directly controlled the systems throughout. If I am not involved in some aspect of the procedure, I cannot certify to its accuracy.
Robert C. Strunk
Sworn to before me and subscribed in my presence this 5th day of December, 1985.
RESUME - [omitted from internet]
PERSONAL INFORMATION - [omitted from internet]
University Graduate School, Cincinnati, OH
Degree: Master of Science in Mathematics (M.S.)
Member: P.M.E. - Honorary Mathematics Society
A.C.M. - Association for Computing Machinery
Activities: Graduate Teaching Assistant (65-67)
Programming Supervisor in the University's Computer Center (66-67)
1961-65: Xavier University, Cincinnati, OH
Degree: Honors, Bachelor of Arts (H.A.B.)
(H.A.B. program requied over 170 hours of classes in a normal eight-semeser carer.)
Majors: Mathematics and Classical Languages
Activities: Band - Military, Marching, and Concert
1957-61: Saint Xavier High School, Cincinnati, OH
Advanced - standing class (Classical Course)
Advanced - placement program (College Courses)
Member: National Honor Society
Letter of Commendation on National Merit Scholarship Exam
Graduated 13th out of 233 seniors
Activities: School Newspaper, Annual, Photoraphy Club, Band
CAREER OBJECTIVE - [omitted from internet]
WORK EXPERIENCE - [omitted from internet]
PERSONAL SKILLS SUMMARY
18+ years experience
Systems Programmer and Applications Programmer
Systems generation, tuning, and modification
Operating Systems: DOS, EDOS, DOS/VSE/MVT, RSTS, VMS, AMOS/L, CP/M
Languages: Four diferent ASSEMBLERS. (Best)
FORTRAN, COBOL, BASIC (Good)
Two different Machine Languages (Good)
RPG, PASCAL, PL/I (Need book)
18+ years experience
Systems: Payroll, AR, AP, GL, inventory, Academic Administrative Systems,
Academic Fund-raising Systems, Multi-User File Access TEchniques,
Combined-file structures, Distributed Computer Systems Computer Hardware:
Have had extensive involvement with the installation of both hardware and software on the following machines:
1) IBM Unit-Record
2) IBM 1520 Computer System
3) IBM 360 Model 30 and Model 50
4) IBM 4341
5) DEC PDP-11/45
6) DEC VAX-11/780
7) Alpha-Micro Computer Systems
8) Communications facilities: dialup, leased, private lines
9) Modems and Multiplexers
10) Synchronous and Asynchronous terminals and printers
11) PBX telephone systems
Seven years experience in the repair of CRT terminals and micro computers
Supervised staff of two to six full-time programmer/analysts (1968-84)
Supervised staf of three to eight student prorammers (1966-76)
Supervised reconstruction of University Band Department quarters (1962-65)
Band president, Senior year in college
Class President, Freshman year in High School
Managed the implementation of all major systems handled by the University's Computer Center since its beginning in 1965. See the above list of projects managed under the heading "Systems Designer". Most of these systems were designed and refined as we progressed through bigger and faster computer systems.
Managed the installation of two IBM 360 Model 30 computer systems, an IBM 360 Model 50 system, an IBM 4341 Group 1 system, a DEC PDP-11/45 system, and a DEC VAX-11/780 system.
Managed the installation of over twenty Alpha-Micro timesharing computer systems, each system having from three to twenty-eight terminals attached.
Managed reconstruction of Xavier University Band's physical plant facilities.
Managed regular publication and delivery of "Edgewood Recorder" for two years.
Teaching: Seven years experience teaching Mathematics and Computer Programming at the college level.
Excellent command of language, diction, and spelling. Wrote hundreds of pages of technical documentation describing the use of specific computer prorams. Was elected to public office.
Wrote 15 editions of "Edgewood Recorder".
Music: Play the Accodion and Drums
Father to three
Accomplished electrician. Have a good understanding o single- and three-phase power, grounding, and power factor load analysis.
RECOUNT OF THE GENERAL
TUESDAY, NOVEMBER 5, 1985
COUNTY OF HAMILTON
STATE OF OHIO
THE BOARD OF ELECTIONS OF HAMILTON COUNTY, OHIO, DOES HEREBY CERTIFY THAT AT THE RECOUNTS HELD ON MONDAY AND TUESDAY NOVEMBER 25TH AND 26TH AND MONDAY DECEMBER 2ND, 1985, INT HE COUNTY OF HAMILTON, THE VOTES CAST FOR THE CANDIDATES, WERE AS FOLLOWS AND APPEARS FROM THE RETURNS FOR ALL THE SELECTEC VOTING PRECINCTS DULY RETURNED TO, OPENED AND CANVASSES, AND NOW REMAINING ON ILEI N THE OICE OF THE BOARD OF ELECTIONS.
WITNESS OUR OFFICIAL SIGNATURES AT CINCINNATI, OHIO, THIS 3RD DAY OF DECEMBER, 1985
EXHIBIT B (PAGE 1 OF 2)
GENERAL ELECTION NOVEMBER 5, 1985
COUNTY OF HAMILTON
STATE OF OHIO
THE BOARD OF ELECTIONS OF HAMILTON COUNTY, OHIO, DOES HEREBY CERTIFY THAT AT THE GENERAL ELECTION HELD ON TUESDAY NOVEMBER 5, 1985 IN THE COUNTY OF HAMILTON, THE VOTES CAST FOR THE CANDIDATES, QUESTIONS AND ISSUES, WERE AS FOLLOWS AND APPEARS ROM THE RETURNS FOR ALL THE VOTING PRECINCTS WITHIN HAMILTON COUNTY, DULY RETURNED TO, OPENED AND CANVASSES, AND NOW REMAINING ON FILE IN THE OFFICE OF THE BOARD OF ELECTIONS.
WITNESS OUR OFFICIAL SIGNATURES AT CINCINNATI, OHIO, THIS 18TH DAY OF NOVEMBER, 1985