Opinion
No. 93-0709.
Oral argument May 25, 1994. —
Decided June 15, 1994.
REVIEW of a decision of the court of appeals. Reversed.
For all petitioners-appellants there was a joint brief by Bruce J. Landgraf, E. Michael McCann, district attorney, Milwaukee County and Fredrick Gordon, Milwaukee and oral argument by Bruce J. Landgraf and E. Michael McCann.
For respondent-respondent-petitioner there were briefs by Catherine M. Canright, Milwaukee and oral argument by Catherine M. Canright and Walter Dickey, co counsel.
For respondent-respondent there was a brief by Edward S. Marion and Murphy Desmond, S.C., Madison.
Amicus curiae brief was filed by Walter Dickey and Legal Assistance Program, Madison for Prof. Walter Dickey and LAIP.
Amicus curiae brief was filed by Richard D. Martin and Office of the State Public Defender, Madison for The State Public Defender.
(Reversing 180 Wis.2d 438, 509 N.W.2d 440 (Ct.App. 1993).)
This is a review of a published decision of the court of appeals, State of Wisconsin, ex rel. Elnora Parker v. Patrick J. Fiedler, 180 Wis.2d 438, 509 N.W.2d 440 (Ct.App. 1993), reversing a judgment of the circuit court for Milwaukee county, Robert W. Landry, circuit judge. The circuit court dismissed the petitions for writs of certiorari, concluding that the Department of Corrections had correctly computed Gerald Turner's mandatory release date. Examining the same statute, the court of appeals concluded that the Department of Corrections had erred.
The court of appeals justifies its computation of the mandatory release date on two grounds, first, that the statute is unambiguous, and second, that the legislative history supports its interpretation. Because neither ground withstands analysis, we reverse the decision of the court of appeals. By no interpretation of the statutory language or the legislative history can this court conclude that the Department has been misinterpreting the statute for 70 years. We therefore conclude, as did the circuit court, that the Department of Corrections computed Gerald Turner's mandatory release date correctly under the applicable statute.
The briefs of both amici, the Legal Assistance to Institutionalized Persons Project (LAIP) and the State Public Defender, raise the specter that if the court of appeals decision is affirmed, numerous prisoners will contest the validity of guilty and no contest pleas entered before 1984 on the ground that when they entered their pleas, they did not understand the potential punishment if convicted. Furthermore, the amici assert that offenders will also have a substantial basis upon which to seek sentence modification.
Turner asserts that retroactive application of the court of appeals' novel interpretation of the statute violates his constitutional rights.
In light of our holding, we need not reach these issues.
From a detached legal perspective this case presents an oft-recurring question of law — what does the statute mean and how should it be applied? However, this case also evokes powerful emotions. Gerald Turner is described in the record as an "unrepentant, apparently homicidal pedophile." This court affirmed Turner's conviction in 1977, and we know well the facts of his appalling crime. We understand the fears of the community that this man, untreated after 18 years behind bars, may hurt another child. We can find no explanation in the record for the Department of Corrections' decision to place Turner in a half-way house in close proximity to an elementary school. This aspect of the Department's decision, while troubling, is not before the court.
In interpreting statutes, we weigh many considerations. The needs of the public and common sense always rank high among them. However, though to many it may appear that public safety and common sense will be served by keeping Turner in prison, the law dictates otherwise in this case. This case illustrates the hard fact that judges sometimes reach decisions that they do not like. We make these decisions "because they are right, right in the sense that the law . . . compel[s] the result." Texas v. Johnson, 491 U.S. 397, 420 (1989) (Kennedy, J. concurring).
This is not an easy case for judges. We are also parents and grandparents. We live alongside the other citizens of this state. But in this country judges cannot tailor their interpretation of the law to fit a particular individual, no matter how heinous his crimes. The integrity of our criminal justice system and of the law itself depends on the courts' consistent application of the same rules to everyone. These principles of the rule of law must guide the decision in this case, as they do all others.
I.
On Halloween night in 1973, nine-year-old Lisa Ann French went trick or treating in a hobo costume in Fond du Lac, Wisconsin. A week later a farmer found her body in a plastic bag. Gerald Turner was convicted of sexually molesting and murdering Lisa Ann. Sections 940.02 (second degree murder), 944.11(1) (indecent behavior with a child), 944.12 (enticing a child for immoral purposes), and 944.17(1) (sexual perversion), Stats. 1973. Turner was sentenced to the Wisconsin State Prisons for an indeterminate term of not more than thirty-eight and one-half years. The Department of Corrections gave him a proposed mandatory release date, which was modified several times over the years because of his misconduct. On October 14, 1992, after serving almost eighteen years, Turner was released from prison and placed in a halfway house in the city of Milwaukee.
Turner v. State, 76 Wis.2d 1, 250 N.W.2d 706 (1977).
Turner was repeatedly denied discretionary parole. He was paroled on October 14, 1992, only because the Department determined that he had reached his mandatory release date. The Department computed Turner's mandatory release date in accordance with its consistent practice of more than 70 years of computing "good time" under sec. 53.11(1), Stats. 1973.
The Department apparently represents that its method of computing good times has been in effect since at least 1917. According to the statutes, the legislature adopted this method of computing good time in 1880.
The Department's computation of good time, Turner's mandatory release date, and Turner's release were challenged in the circuit court for Milwaukee county by three citizen relators (one of whom later withdrew) and two public officials. We shall refer to the remaining four as the relators. Although the Department of Corrections contested the relators' action in the circuit court and court of appeals, the Department withdrew from participation after filing a brief in this court opposing the decision of the court of appeals. The relators proceeded in circuit court by writ of certiorari.
The Department of Corrections was represented by attorney Edward Marion, of Murphy Desmond, S.C., in the circuit court and court of appeals. After Attorney Marion filed the Department's brief in this court seeking reversal of the decision of the court of appeals, he wrote the court on behalf of the Department of Corrections requesting permission to withdraw the Department's brief and advising the court that the Department "will not participate further in this appeal." Letter dated April 24, 1994, on file with the Clerk of Supreme Court. This court issued an order denying the Department's request to withdraw the brief but honoring the Department's request to waive participation in further proceedings. Order dated May 13, 1994, on file with the Clerk of Supreme Court. Accordingly when the court refers to the Department in this opinion, the reference is to its brief before this court or to the record in the circuit court.
The Department's brief asserts that certiorari was not the appropriate remedy in this case because the grant of a mandatory release is a ministerial, not a discretionary, act.
Relying on Richards v. Young, 150 Wis.2d 549, 441 N.W.2d 743 (1989), Turner and the Department assert that the circuit court lacked subject matter jurisdiction (or "competency," to use the terminology of some recent cases), sec. 801.04, Stats. 1991-92, because the relators failed to comply with sec. 227.40, Stats. 1991-92.
Section 227.40(1) states that "the exclusive means of judicial review of the validity of a rule shall be an action for declaratory judgment as to the validity of such rule brought in the circuit court for Dane county." This action was not commenced as a declaratory judgment or in Dane county.
Section 227.40(5) states that "the joint committee for review of administrative rules [sec. 13.56, Stats. 1991-92] shall be served with a copy of the petition" in the declaratory judgment action seeking judicial review of the validity of a rule. The joint committee was not served.
In response to the Department's and Turner's arguments, the relators assert that the controversy does not involve the validity of an administrative rule and that if it does, this proceeding falls within the exception provided under sec. 227.40(2) (a), namely "a civil proceeding by a state officer to enforce a statute."
Although we do not necessarily accept the relators' arguments that they complied with sec. 227.40, we assume, for purposes of analysis only, that the relators have complied with sec. 227.40. As the Department's brief acknowledges, p. 19, n. 10, "this court did not accept review of the court of appeals' decision only to avoid the merits by reversing on a procedural error. . . . The Department does not ask this court to reverse solely on these grounds."
Elnora Parker, one of the citizen relators, lives "within blocks" of the half-way house facility to which Turner was released; Carol A. Arendt, the other citizen relator, has a daughter who attends the school which is "adjacent to the back door of" the half-way house. Their petition complains that Turner's release "at a time prior to his rehabilitation and prior to his Mandatory Release date, constitutes substantial damage [to them], who as citizens of the State of Wisconsin are entitled to protection from unreformed criminals until they are required by law to be released from imprisonment."
The other two relators are E. Michael McCann, the elected district attorney for Milwaukee county whose function is to "prosecute all criminal actions" within his jurisdiction, sec. 978.05(1), Stats. 1991-92, and Frederick G. Gordon, an elected alderman of the Milwaukee city council, who represents the area in which Turner's half-way house is located.
Turner's and the Department's briefs raise the question of whether the relators have standing to challenge Turner's release in the circuit court, the court of appeals, or this court.
The law of standing is complex and depends in large measure on the type of claim asserted. The purpose of the requirement of standing is to ensure that a concrete case informs the court of the consequences of its decision and that people who are directly concerned and are truly adverse will genuinely present opposing positions to the court. We acknowledge that the court is presented with a concrete case and that the opposing positions have been forcefully briefed. Although we do not necessarily accept the relators' arguments that they have standing, even when standing is liberally construed, we assume, for purposes of analysis only, that the relators have standing.
II.
The primary issue presented in this case is one of statutory interpretation. The court must determine how good time is computed under sec. 53.11, Stats. 1973. This 1973 statute was in effect when Turner was sentenced, and it remained relatively unchanged until 1983. The 1983 law is not at issue in this case.
We do not discuss other good-time credits or forfeiture of good time.
Effective for crimes committed on or after June 1, 1984, a prisoner's mandatory-release date was set at "two-thirds of the sentence." See 1983 Wis. Act 66, 528; sec. 302.11(1) and (9), Stats. 1991-92.
All the parties agree that for several decades, at least for more than 70 years, the Department of Corrections has consistently construed sec. 53.11 in the same way. The court of appeals concluded, however, that the Department's interpretation contravenes the unambiguous words of the statute.
Statutory interpretation is a question of law which this court decides, benefitting from the decisions of the circuit court and the court of appeals. The principal objective of statutory interpretation is to ascertain and give effect to the intent of the legislature. Green Bay Development Authority v. Bee Frank, Inc., 120 Wis.2d 402, 409, 355 N.W.2d 240 (1984).
Our first step is, of course, to examine the statutory language itself. Section 53.11, Stats. 1973, provides that the warden shall keep a record of an inmate's conduct and that each inmate who comports with the statute is entitled to good time or diminution of sentence as follows: First year, 1 month; second year, 2 months; third year, 3 months; fourth year, 4 months; fifth year, 5 months; every year thereafter, 6 months. Section 53.11(1), Stats. 1973, reads as follows:
Section 53.11(1). The warden or superintendent shall keep a record of the conduct of each inmate, specifying each infraction of the rules. Each inmate who shall conduct himself in a proper manner and perform all the duties required of him shall be entitled to good time or diminution of sentence according to the following table, prorated for any part of a year: First year, one month; second year, 2 months; third year, 3 months; fourth year 4 months; fifth year, 5 months; every year thereafter, 6 months.
Reading sec. 53.11(1) with secs. 53.11(1)(6) and (7)(a), the court of appeals' opinion steadfastly maintains that the statute is unambiguous and can be interpreted in only one way. This contention does not withstand analytical scrutiny.
Section 53.11(6), Stats. 1973, provides as follows: "Allowances for good conduct earned in any institution shall be allowed in the institution to which an inmate may be transferred."
Section 53.11(7)(a), Stats. 1973, provides as follows: "An inmate or parolee having served the term for which he has been sentenced for a crime committed after May 27, 1951, less good time earned under this chapter and not forfeited as herein provided, shall be released on parole. . . ."
The court of appeals' opinion seems internally inconsistent. On the one hand, it asserts that the statute unambiguously states how to compute good time and thus finds it unnecessary to consult either legislative history or the administrative interpretation as aids to statutory interpretation. On the other hand, in responding to Turner's argument that applying the court of appeals' method of computing good time retroactively would violate his constitutional rights, the court of appeals consults selected legislative history to demonstrate to its satisfaction that its judicial interpretation of the statute was foreseeable and "merely confirms what the statute has clearly required, at least [since the 1919 law]." State ex rel. Elnora Parker, 180 Wis.2d at 465.
First, the key statutory provision at issue, sec. 53.11(1), is on its face flawed. Although this provision refers the reader to a computation of good time set forth in a "table," no graphic representation in tabular form exists. Albeit on the surface it seems a minor issue, this obvious infirmity raises doubts about the clarity of the statute. The significance of the table will become apparent when we discuss the legislative history.
Furthermore, no matter how sec. 53.11(1) or sec. 53.11(6) and (7) are parsed, they do not describe how to compute good time. It is fair to say that sec. 53.11 poses something of an interpretive riddle. Comments about the difficulty in interpreting and applying sec. 53.11 appeared in the legal literature long before this case arose. The statute has been characterized as "complex" and "open to several interpretations." Courts in other states faced with interpreting similar statutes have acknowledged that the statutes are subject to at least two interpretations.
The court has observed that "in what is probably an understatement, the computation of good time credits and the administration of secs. 53.11(7)(a) and (7)(b) have been characterized as `complex.'" State ex rel. Hauser v. Carballo, 82 Wis.2d 51, 58, n. 13, 261 N.W.2d 133 (1978), quoting Rice v. Schmidt, 277 F. Supp. 811, 814 (E.D. Wis. 1967).
Manual, Legal Assistance to Inmates and Mental Patients, sec. 6.02 (Univ. of Wis. Law School 1974).
See Ex Parte Blocker, 193 P. 546, 547 (Colo. 1920) (adopting the Department's method of computation); Oakland Prosecutor v. Department of Corrections, 503 N.W.2d 465, 469 (Mich.App. 1993), leave to appeal denied, 508 N.W.2d 498 (adopting the Department's method of computation which is referred to as the "immediate usage method"; relators' method referred to as the "savings account method").
By no stretch of the statutory language or the imagination can this court conclude, as the court of appeals did, that the words of sec. 53.11(1), whether read alone or together with other subsections of sec. 53.11, are subject to only one interpretation.
Section 53.11(1) is open to at least two possible interpretations. The statutory words "first year," "second year," "third year," "fourth year" and "fifth year" can be interpreted as either the year of the sentence imposed or the year of the sentence actually served in prison. Both interpretations fulfill the legislature's objective of protecting the safety of correctional officers and inmates by encouraging prisoners to behave. The Department, Turner and the circuit court advance the first interpretation, and the relators and the court of appeals promote the second.
Under the Department's interpretation, a prisoner earns 1 month credit after 11 months of his first year in prison. For example, a person who behaves properly for the first 11 months in prison earns 1 month of good time and the prison portion of that first year of the sentence is diminished by 1 month; the second year of that prisoner's sentence begins at the end of 11 months. The phrase "diminution of sentence" in sec. 53.11(1) provides support for this result. "Diminution" means "to make smaller." Nothing in the language of the statute contravenes this interpretation. Nothing in sec. 53.11(1) requires or implies that a person must serve a full year to earn the good time allowed under the statute.
The relators describe this computation as a mathematical fiction. They object to viewing a year as completed when 11 months are served the first year, when 10 months are served the second year, when 9 months are served the third year and so forth.
Under the court of appeals' interpretation, a prisoner earns 1 month credit after serving the first 12 months in prison. The court of appeals asserts that no good time can be credited until the prisoner "earns" it. The word "earned" does not appear in sec. 53.11(1); the court of appeals imports the word from secs. 53.11(6) and (7), which were enacted after the adoption of sec. 53.11(1).
The difference between the computations of the court of appeals and the Department becomes obvious merely by looking at the computation of good time in the first year. According to the court of appeals and the relators, the prisoner must serve 1 year (or 365 days) before the prisoner is entitled to 1 month (or 30 days) of good-time credit. According to the Department, the prisoner must serve 11 months before the prisoner is entitled to 1 month of good time.
Section 53.11(7)was adopted in 1951, not to change the computation of good-time credit but to ensure that a prisoner who was released under sec. 53.11(1) on the mandatory release date was placed on parole rather than finally discharged. Laws of 1951, ch. 256.
The word "earned" appears, however, in Laws of 1880, ch. 238, sec. 2; sec. 4928a 2., 1889 Sanborn Berryman's Annot. Stats.
The word "earned" is the focal point of the court of appeals' decision. Relying on the dictionary definition of "earned," the court of appeals concludes that no one can "earn" anything prospectively. Thus, according to the court of appeals, sec. 53.11(1), when read in conjunction with sec. 53.11(7), means that the "prisoner does not `earn' the first month of good time until the prisoner has completed the first year of confinement, and does not `earn' the 2 additional months until the prisoner has completed the second year of confinement, and so on". . . . State of Wisconsin ex rel. Elnora Parker v. Patrick J. Fiedler, 180 Wis.2d at 460.
The court of appeals gives the following example of how to calculate the time to be served for a defendant sentenced to a 5-year term of incarceration. "If, during the first year of confinement, the prisoner conducts him or her-self `in a proper manner and perform[s] all the duties required,' the prisoner will have earned `good time or diminution of sentence' of one month for that first year. . . . Thus, after completion of that first year, the prisoner will have three years and eleven months left to serve (the five-year sentence, less the one year served plus the one month of good time earned during the first year). . . . At the end of the fourth year of confinement, the prisoner who has earned good time under section 53.11 for that year will have to serve two months (the five year sentence, less the four years already served plus the one month of good time earned for the first year, the two months of good time earned the second year, three months good time earned for the third year, and four months good time earned for the fourth year). Under section 53.11, the prisoner is then entitled to have earned good-time credit prorated for the final two months at the rate for the fifth year. . . . By contrast, the [Department's computation] . . . awards the prisoner nine months good-time credit for the fourth and fifth years, even though the prisoner never serves any part of those years, and accordingly could not have `conducted himself in a proper manner and perform[ed] all the duties required of him. . . ." State ex rel. Elnora Parker, 180 Wis.2d at 461, n. 7. The prisoner under the court of appeals' calculation gets 10 months good-time credit.
The relators' brief at 7-8 in this court presents a somewhat different computation than the one presented by the court of appeals. The relators converts "years" in prison into "days," even though the statute speaks of good-time credit in terms of months and years.
The Department's brief asserts that while the relators' computation in terms of days is technically feasible today with personal computers, in the early part of this century the legislature could not have intended this complex computation resulting in parts of days.
The court of appeals' computation should be compared to the Department's example of how to calculate the time to be served for a defendant sentenced to a 5-year term of incarceration: "An inmate with a single five-year sentence which had a beginning date of 5-16-74 has a projected discharge date of 5-16-79. Such a person may earn one year, three months of statutory good-time pursuant to Sec. 53.11, Stats. . . . ." Section DOC 302.21, Appendix, Wis. Admin. Code. This computation is the same as the third column in the 1880-1917 statutory table. See pp. 18 and 19.
We agree with the court of appeals that the legislature intended a prisoner to earn good time. Indeed, no one in this case disputes this conclusion. The issue remains, however, how much time the prisoner must serve in prison before the statutorily specified amount of good time is earned. The legislature could have chosen to give a prisoner 1 month credit for 11 months served or 1 month credit for 12 months served. Put another way, the court of appeals does not tell us why a prisoner has to serve 12 months, rather than 11 months, to "earn" the 1 month of good time. The court of appeals does not explain why the statutory language unambiguously leads it to the conclusion that the legislature decided that a prisoner has to serve 12 months to earn 1 month good-time credit.
The Department's brief notes that because good time credit may be forfeited for misconduct, good time is as much "earned" under the Department's interpretation of the statute as it is earned under the court of appeals' interpretation. Department's Brief at 31, n. 15.
The Michigan court of appeals similarly dismissed the argument that the Department's method of computation did not require the prisoner to earn good time, writing as follows: "Moreover, contrary to plaintiffs assertion, it is simply not true that `prisoners are routinely given a windfall of credit for good time [under the Department's method] before a day is served instead of earning credit by demonstrated good behavior.' As indicated above, prisoners are provided with a good-time credit computation sheet upon entering prison. These computations are only projections and are used to induce good behavior, because a prisoner must remain free of misconduct in a given month in order to earn the good time credit. . . ." Oakland Prosecutor v. Department of Corrections, 503 N.W.2d 465, 470 (Mich.App. 1993).
The court of appeals' decision simply states that "the Department . . . has, for many years, been awarding the good time before it is earned. . . ." State ex rel. Elnora Parker, 180 Wis.2d at 458.
While the court of appeals' interpretation of sec. 53.11(1) appears reasonable at first blush, we must straightaway point out that it is problematic. The court of appeals' interpretation, unlike the Department's, does not fit all sentences a court might impose.
Under the court of appeals' interpretation, a person sentenced to 1 year in prison would receive no good-time credit. Yet the statute expressly addresses good-time credit relating to the first year. The Department's interpretation of sec. 53.11(1)would credit such a prisoner with 1 month good time. This puzzle is not squarely addressed by either the court of appeals or the relators. The court of appeals suggests how the Department could prorate good time during a prisoner's last year, but that explanation does not solve the problem of a 1 year sentence.
Section 53.11(1) provides proration for only part of a year.
The court of appeals' interpretation also denies good-time credit to other persons who appear entitled to credit under sec. 53.11(1). The amicus brief of the University of Wisconsin Law School Legal Assistance to Institutionalized Persons Project (LAIP), points out that when prisoners are left with between 12 and 18 months to serve until mandatory release, the court of appeals' computation denies them their full entitlement to good time. The amicus's computations demonstrating this quandary are set forth in the margin.
One of the authors of the brief is Walter Dickey, Professor of Law, University of Wisconsin, who was the principal drafter of the Department's rulemaking project and Administrator of the Division of Corrections of the State of Wisconsin Department of Health and Social Services from 1983-1987.
The amicus brief at p. 3 gives the following example of what happens when the court of appeals' calculation is applied to a defendant sentenced to a 6-year term of incarceration. After serving 4 years of the 6-year sentence, according to the court of appeals calculations set forth in the margin at note — above, the prisoner will have 1 year and 2 months left to serve (the 6-year sentence less the 4 years already served plus the 1 month of good time earned for the first year, the 2 months for the second year, the 3 months for the third year, and the 4 months for the fourth year).
The problem occurs here, according to the amicus. The prisoner now has 14 months to serve before reaching his mandatory release date. Under the court of appeals' decision, the prisoner must serve 12 of these 14 months in prison before he earns the 5 months allowed for his fifth year under the statute. But the prisoner does not receive the 5 months good time because after serving the fifth year he has only 2 months left to serve. This would not be a problem if good time could be prorated during the fifth year. But good time cannot be prorated during the fifth year, because sec. 53.11(1) expressly states that good time is prorated only for "part" of a year. The 14 months left to be served are not a "part" of the year.
To determine which computation of good-time credit the legislature intended, we examine the legislative history of the statute and the administrative agency's interpretation of the statute, two traditional indicia of legislative intent.
III.
The legislative history of sec. 53.11 (including the portions of legislative history relied upon by the court of appeals) overwhelmingly supports the Department's method of computing good time, not that of the court of appeals. Indeed, for a statute this old, it is unusual to unearth so much legislative history and such clear expressions of legislative intent directly applicable to the issue presented to the court.
The first good conduct statute was enacted in 1860. It provided that the conduct of each prisoner was to be recorded, that the prisoner was to receive a certificate of good conduct, and that the warden may diminish the term of sentence not more than 5 days for each certificate. The statute provides as follows:
"It shall be the duty of the State Prison Commissioner to require his deputy to keep a true record of the conduct of each convict, specifying each infraction of the rules of discipline. At the end of each month the said deputy shall give a certificate of good conduct to each convict who shall require it, against whom is recorded no infraction of the rules of discipline; and, upon each such certificate, the commissioner may, at his discretion, diminish the term of any convict sentenced for a specific time, not more than five days. All such certificates shall remain on file in the prison office, subject at any time to be annulled by the Governor for subsequent misconduct of the convict." Laws of 1860, ch. 324, sec. 1, as amended Laws of 1861, ch. 122, sec. 1. (Emphasis added.)
This provision was recreated in substantially the same language by Laws of 1871, ch. 115, sec. 51, and Laws of 1873, ch. 193, sec. 48. This provision appears as sec. 4928 in the 1889 Sanborn Berryman's Annot. Stats.
The phrase "at the end of each month" indicates that the prisoner had to serve a full month in order to get 5 days off the sentence. The words "and upon" indicate that the award of 5 days was conditioned upon the receipt of the certificate.
In 1880 the legislature adopted a second provision relating to good conduct. The new provision did not replace, but merely supplemented, the 1860 law. The new 1880 provision, setting forth a "table" showing good time "made" and time "to be made," stated:
Laws of 1880, ch. 238, sec. 1. This provision appears as sec. 4928a in the 1889 Sanborn Berryman's Annot. Stats.
It is unclear how the two statutes related to each other and how the 5-day reduction for each 30 days fits with the computations in the Table.
Credit for good conduct; forfeiture for bad; transportation on discharge. Section 4928a [Ch. 238, 1880.] 1. Every convict who is now or may be hereafter confined in the state prison, and shall conduct himself in a peaceful and obedient manner, and faithfully perform all the duties required of him, shall be entitled to a diminution of time from the term of his sentence, not exceeding the amounts specified in the following table, for the respective years of his sentence, and pro rata for any part of a year, where the sentence is for more than a year:------------------------------------------------------------------------ Year of Good Time Total Good Time Time To Be Made If Full Sentence. Granted. Made. Good Time Is Made. ------------------------------------------------------------------------ First year One month One month Eleven months Second year Two months Three months One year and nine months Third year Three months Six months Two years and six months Fourth year Four months Ten months Three years and two months Fifth year Five months One year and three months Three years and nine months Sixth year Six months One year and nine months Four years and three months Seventh year Six months Two years and three months Four years and nine months ------------------------------------------------------------------------
Where the sentence exceeds seven years, for every year after the seventh, if the conduct of the prisoner continues to correspond with the requirements of this act, he shall receive the same diminution as provided for the sixth and seventh years.
The last column labeled "Time To Be Made If Full Good Time is Made," represents the difference between the year of sentence (column 1) and the cumulative good time (column 3) and is obviously the term a prisoner sentenced to the number of years in column 1 is to serve before release. The Department's computations comport with this table. See the Department's calculation of a 5-year term in the margin at note 19.
In 1898 the legislature combined the 1860 and 1880 provisions into one section, numbered sec. 4928, relating to credit for good conduct. The 1898 statute reads, in part, as follows:
Section 4928. The deputy warden shall keep a true record of the conduct of each convict, specifying each infraction of the rules of discipline. At the end of each month, the said deputy shall give a certificate of good conduct to each convict who shall require it, against whom is recorded no infraction of the rules of discipline. Every convict who is now or may be hereafter confined in the state prison and shall conduct himself in a peaceful and obedient manner and faithfully perform all the duties required of him shall be entitled to diminution of time from the term of his sentence, not exceeding the amounts specified in the following table, for the respective years of his sentence and pro rata for any part of a year, where sentence is for more than a year:------------------------------------------------------------------------ Year of Good Time Total Good Time Time To Be Made If Full Sentence. Granted. Made. Good Time Is Made. ------------------------------------------------------------------------ First year One month One month Eleven months Second year Two months Three months One year and nine months Third year Three months Six months Two years and six months Fourth year Four months Ten months Three years and two months Fifth year Five months One year and three months Three years and nine months Sixth year Six months One year and nine months Four years and three months Seventh year Six months Two years and three months Four years and nine months ------------------------------------------------------------------------ Where the sentence exceeds seven years, for every year after the seventh, if the conduct of the prisoner continues to correspond with the requirements of this section, he shall receive the same diminution as provided for the sixth and seventh years. . . .
While the 1898 version of the good conduct statute retains a reference to an "end of the month certificate" for good conduct (which appeared in the 1860 law), this version no longer connects the certificate to any reduction in sentence. The purpose of the certificate under the 1898 statute is not stated and remains unclear.
Aside from the change in the certificate, the 1898 provision is substantially similar to the 1880 law and reprints the 1880 table to which the Department's computations conform. This table appears in the statutes from 1880 until 1919. The court of appeals, State of Wisconsin ex rel. Elnora Parker v. Patrick J. Fiedler, 180 Wis.2d at 465, and the attorney for the citizen relators have acknowledged, as they must, that this table supports the Department's interpretation of sec. 53.11(1).
When the 1898-1917 good conduct statute, sec. 4928, was renumbered sec. 53.11 in 1919, the text of the 1898 law remained unchanged. The last two columns of the table were, however, deleted.
Laws of 1919, ch. 348, sec. 13, renumbers and amends sec. 4928 to read as follows:
Section 13. Section 53.11 Credit for good conduct; forfeiture for bad. * * * (1) The deputy warden shall keep a true record of the conduct of each convict, specifying each infraction of the rules of discipline. At the end of each month the said deputy shall give a certificate of good conduct to each convict who shall require it, against whom is recorded no infraction of the rules of discipline. Every convict who is now or may be hereafter confined in the state prison and shall conduct himself in a peaceful and obedient manner and faithfully perform all the duties required of him shall be entitled to a diminution of time from the term of his sentence, not exceeding the amounts specified in the following table, for the respective years of his sentence and pro rata for any part of a year, where the sentence is for more than a year:
---------------------------------------------------------- Year of Good time * * * * * * sentence. granted. * * * * * * ---------------------------------------------------------- First year One month * * * * * * Second year Two months * * * * * * Third year Three months * * * * * * Fourth year Four months * * * * * * Fifth year Five months * * * * * * * * * Six months * * * * * * Every year * * * * * * thereafter * * * * * * * * * * * *
The court of appeals' concludes that this 1919 deletion of part of the table demonstrates the legislature's intent to change the method of computation that had been expressed in the table from 1880 until 1919. State ex rel. Elnora Parker, 180 Wis.2d at 465.
Counsel for the citizen relators has taken two inconsistent positions before this court about the 1919 statute: Counsel argues at one point that the 1919 legislature intended to repeal the earlier version of the table and change the method of computing good time. In a supplemental letter submitted at oral argument, counsel argues that there is "utterly no indication in the 1919 law that the 1917 method of calculation was intended to be changed."
We conclude that counsel's latter position is correct. First, the language of the 1919 statute relating to good time is essentially the same as that of the 1880 statute. When the legislature retains the same statutory language it has used for nearly 40 years, it is not reasonable to infer from the mere deletion of part of a table that the legislature is changing its nearly 40-year-old method of computing good time. Second, nothing in the legislative history reveals a legislative intent to change the method of computation. Indeed the legislative history shows that the legislature did not intend to make a change in the computation of good time. The 1919 revision was part of a Revisor's Bill, indicating that it was drafted to simplify, clarify and consolidate provisions relating to state prisons and place them into chapter 53 of the statutes. Because revisor's bills generally contain little or no substantive changes, the court would not expect the legislature to enact a major revision of the method of computing good time in such a bill. But we need not base our conclusion that the legislature did not intend to change the method of computing good time simply on the nature of revisor's bills or on conjecture or logical deduction. The revisor's drafting note to sec. 53.11(1) clearly explains the purpose of the 1919 revision of the 1917 law and the deletion of part of the table as follows: The "matters stricken in subsection (1) are merely argumentative, and surplusage. . . ."
Senate Bill No. 54, S., 1919 Legislature. This bill became Laws of 1919, ch. 348, effective June 16, 1919.
Citing to the Cyclopedic Law Dictionary of 1922 (authored by James Cahill, published by Chicago, Callaghan and Company), Turner's supplemental brief points out that the word "argumentative" was defined as "indirect; by inference," a somewhat different meaning than the word now has.
The court of appeals' reasoning that the 1919 law changed the method of computing good time is faulty because the court of appeals examined only selective parts of the legislative history. The court of appeals failed to understand the nature of the 1919 Revisor's Bill and failed to put the 1919 Bill into historical perspective. The court of appeals reasoned that retention in 1919 of the express requirement of an "end of each month certificate" and deletion of part of the table add up to a rejection of the Department's method of computation. State ex rel. Elnora Parker, 180 Wis.2d at 464-65. Apparently the court of appeals did not consider that the legislature had retained both "the end of the month certificate" and the table from 1880 to 1919 and that the legislature evidently had not seen any conflict between the two provisions for nearly 40 years.
The relators argue and the court of appeals concluded that the certificate requirement is inconsistent with the notion that a warden may deem a year to be less than a year. They seem to take the position that the "certificate" requirement trumps the table.
As the court of appeals concedes, its method of computation, in contrast to the Department's method, does not conform to the 1880-1917 statutory table. State ex rel. Elnora Parker, 180 Wis.2d at 465. The relators' present a table in their supplemental letter brief which they claim conforms to the 1917 statutory table. Although the first four columns of their table are expressed in terms of days rather than months and years, they seem to conform to the 1917 table. But as the relators concede, their last column — the one stating the total amount of time remaining to be served — "differ[s] from the 1917 table." In reality, the relators' method of computing good time compels them to present a table specific to Turner, whereas the statutes present a table which can be used for any prisoner regardless of sentence.
The legislative history does not end in 1919. Section 53.11(1) was amended in 1947 and then remained unchanged until 1983. Section 53.11(1) read as follows from 1947 to 1983:
Laws of 1947, ch. 519, sec. 1.
"The warden or superintendent shall keep a record of the conduct of each inmate, specifying each infraction of the rules. Each inmate who shall conduct himself in a proper manner and perform all the duties required of him shall be entitled to good time or diminution of sentence according to the following table, pro-rated for any part of a year: First year, one month; second year, 2 months; third year, 3 months; fourth year, 4 months; fifth year, 5 months; every year thereafter, 6 months."
Thus, as of 1947, the statute no longer refers to the prisoner receiving a certificate. This deletion was long in coming since the monthly certificate did not appear to have any purpose after 1898. We do not find in the legislative history any reason for the deletion of the certificate.
The Joint Interim Committee on Public Welfare Laws introduced the 1947 law amending sec. 53.11. The Committee's comment appended to sec. 53.11, Stats. 1947, does not explain the deletion of the certificate. It is apparent from the comment, however, that no substantive change was contemplated.
The 1947 statute, however, continues to refer to "a table." This reference is apparently a carryover from the 1880 through 1945 statutes. No tabular representation of the computation is set forth in the 1947 statute. Rather, the 1947 statute replaces the earlier table format with a narrative description of the first two columns of the table which was part of the statutes from 1880 until 1947.
A final piece of legislative history is the 1983 revision of the good-time credit. The legislative history of the 1983 law lends support to the Department's method of computing good time.
The relators assert that in 1983 the legislature intended to ensure that the intent of the 1919 revision be followed. Thus, urge the relators, the legislature put an end to the Department's tortured method of computing good time and mandated that the mandatory release date be calculated as two-thirds of the sentence. According to the relators, the 1983 two-thirds calculation results in approximately the same mandatory release date as their computation of Turner's mandatory release date. Brief at 7-8, 19-20. We could find no support in the legislative history of the 1983 revision for the relators' view of the legislative intent.
1983 Wis. Act 66, sec. 1, amended sec. 53.11(1) to read as follows:
The warden or superintendent shall keep a record of the conduct of each inmate, specifying each infraction of the rules. Each inmate who acts properly and performs all required duties is entitled to good time or diminution of sentence at a rate of one day for each 2 days the inmate serves.
1983 Wis. Act 528, sec. 2, amended sec. 53.11(1) to read as follows:
The warden or superintendent shall keep a record of the conduct of each inmate, specifying each infraction of the rules. Except as provided in subs. (1m) and (7), each inmate is entitled to mandatory release on parole by the department. The mandatory release date is established at two-thirds of the sentence. Any calculations under this subsection or sub. (2)(b) resulting in fractions of a day shall be rounded in the inmate's favor to a whole day.
In contrast, the amicus LAIP asserts that the legislative intent in 1983 was to require prisoners to serve more time before mandatory release than was required under the Department's computation of good time. The legislative history demonstrates that the legislature was aware that the 1983 change extended the mandatory release date of persons sentenced to more than 4 years. See Fiscal Estimate, Assembly Bill AB 1011, 1983 Session (Dept. of Health and Social Services, 2/9/84), in the drafting file for 1983 Wis. Act 528. The amicus's brief explains that the court of appeals' computation of sentences under the pre-1983 law may in some instances actually result in longer sentences than those under the 1983 law.
The Fiscal Note states in part: "Since persons sentenced to 4 years or less will have a shorter mandatory release date under the revised formula we assume that these offenders will want to have the new law applied to them. Persons sentenced to more than 4 years will have longer mandatory release dates and will most likely not request that their sentences be computed under the new system."
The LAIP brief(p. 4) gives the following illustration: "For example, on a 20-year sentence, one reaches his mandatory release date in 11 years, 3 months under the Department's interpretation of [the pre-1983 law]. Under the [1983 law] one reaches his mandatory release date in 13 years, 4 months. Under the [pre-1983 law] as interpreted by the court of appeals, one reaches his mandatory release date after serving approximately 15 years."
The legislature granted to persons serving sentences under the old law an opportunity to "opt-in" to the new law. We agree with the LAIP amicus brief that this "opt-in" provision further demonstrates the legislature's understanding that the computations under the 1983 law would be different from those under the prior law. See Fiscal Estimate, Assembly Bill AB 1011, 1983 Session (Dept. of Health and Social Services, 2/9/84), in the drafting file for 1983 Wis. Act 528. Turner apparently opted to serve under the old law.
To summarize, the legislature has revised sec. 53.11 several times since its forerunner was adopted in 1860. Read as a whole, the legislative history evidences that the legislature set forth its method of computing good time in 1880 and did not change the method until 1983. The legislature made clear, as early as 1880, that a prisoner who behaved well would earn 1 month after serving 11 months of the first year sentence. Although the certificate requirement does not dovetail with the legislature's table, the legislative history demonstrates the certificate to be an historical relic which disappears in 1947. However, the table in the statutes from 1880 until 1919 table, which supports the Department's method of computation, remains in the statute until 1983, albeit in condensed or narrative form.
IV.
In addition to examining legislative history to determine legislative intent, a court looks to the interpretation of the statute by the administrative agency charged with its enforcement. The interpretation of a statute presents a question of law, and the "black-letter" rule is that a court is not bound by an agency's interpretation. Nevertheless courts frequently refrain from substituting their interpretation of a statute for that of the agency charged with the administration of a law. As the court of appeals correctly acknowledged, State ex rel. Elnora Parker, 180 Wis.2d at 459, courts frequently give deference to the interpretation of statutes by administrative agencies charged with their enforcement.
Courts will give varying degrees of deference to an agency's interpretation of a statute when they have concluded (1) that the legislature charged the agency with the duty of administering the statute; (2) that the agency's interpretation is of long-standing; (3) that the agency's interpretation employs its expertise, technical competence and specialized knowledge; and (4) that through interpretation and application of the statute, the agency can provide uniformity and consistency in the field of its specialized knowledge. Lisney v. LIRC, 171 Wis.2d 499, 505, 493 N.W.2d 14 (1992). The weight that is due an agency's interpretation of the law depends on the comparative institutional capabilities and qualifications of the court and the administrative agency.
A court does not, however, give deference to an agency's interpretation of a statute when the court concludes that the agency's interpretation directly contravenes the words of the statute, is clearly contrary to legislative intent, or is otherwise unreasonable or without rational basis. Lisney v. LIRC, 171 Wis.2d 499, 506, 493 N.W.2d 14 (1992). Because it concluded that the agency interpreted an unambiguous statute at variance with the words of the statute, the court of appeals did not give deference to the Department's interpretation in this case. State ex rel. Elnora Parker, 180 Wis.2d at 459, 461-62.
An agency's interpretation of a statute is reasonable if it accords with the language of the statute, the statute's legislative history and the legislative intent and if the interpretation is consistent with the statute read as a whole, and the purpose of the statute; and if the interpretation is consistent with judicial analyses of the statute. Lisney v. LIRC, 171 Wis.2d 499, 507, 493 N.W.2d 14 (1992).
The Department's interpretation of sec. 53.11 has existed for more than 70 years, and, unlike many legislative enactments or administrative policies or rules, the Department's method of computing good time has been widely disseminated. Judges, prosecutors, defense counsel, and thousands of prisoners have long been familiar with the Department's computations. According to the court of appeals, the Department has been in error for decades but no one has noticed until now. This is unlikely in light of the statute's repeated use.
Furthermore, the Department's interpretation has consistently been validated by the legislature, the courts and commentators.
Despite the fact that sec. 53.11 has been revised a number of times, the legislature did not amend sec. 53.11(1) until 1983 to require the Department to compute good time in a different manner. The history of the Department's rulemaking also permits the inference that the legislature acquiesced in the Department's computations of good time.
Prior to 1978, the legislature delegated correctional policymaking to correctional professionals through an exemption from rulemaking requirements. Thereafter the legislature required the Department to promulgate administrative rules. Professor Walter Dickey of the University of Wisconsin School of Law was the principal drafter of the Department's rules. According to Professor Dickey, the Department's lengthy rule-drafting project involved careful analysis of statutory and case law requirements and, where necessary, the rules brought agency policy into compliance with the statute. Professor Dickey recounts that the legislature was kept informed during the entire rulemaking process and that the Legislative Council reviewed drafts before they went to the legislature's standing committee. The legislature could have, but did not, veto any of the corrections rules.
Professor Dickey describes the formulation of the rules in his article entitled The Promise and Problems of Rule-making in Corrections: The Wisconsin Experience, 1983 Wis. L. Rev. 285.
In 1992 when Turner was released, an appendix to Department rule, DOC 302.21 (3)(a), set forth the computation of good time and mandatory release under sec. 53.11 (1). See also Register, Aug. 1979, No. 284, Corrections, Appendix Note HSS 320.21.
Walter Dickey, The Promise and Problems of Rule-making in Corrections: The Wisconsin Experience, 1983 Wis. L. Rev. 285, 320.
Walter Dickey, The Promise and Problems of Rule-making in Corrections: The Wisconsin Experience, 1983 Wis. L. Rev. 285, 304-305.
Professor Dickey submitted an affidavit in the circuit court in support of the method of computation used by the Department of Corrections in this case. In addition to reiterating his published observations, Professor Dickey stated that the rulemaking process was overseen, in part, by an advisory committee which included prosecutors and public defenders and that no one questioned the correctness of the Department's computations of good time and mandatory release.
Although neither this court nor the court of appeals has previously considered a challenge to the Department's computation of good time, the Department's method of computing good time has been relied upon by this court. In State ex rel. Hauser v. Carballo, 82 Wis.2d 51, 261 N.W.2d 133 (1978), mandatory and discretionary parolees challenged the Department's procedures in determining the forfeiture of a parolee's good-time credits. In ruling on this issue the court relied on the Department's computations for crediting good time as explained in a 1974 Manual entitled Legal Assistance to Inmates and Mental Patients prepared by Professor Frank Remington, Attorney James Glover and Professor Walter Dickey for the Legal Assistance to Institutionalized Persons Project (LAIP) of the University of Wisconsin Law School. The court stated that good-time credit is granted at the commencement of the prisoner's sentence and the mandatory release date is "calculated by subtracting from the maximum term to which the prisoner was sentenced the entire credit under sec. 53.11 (1) and 53.12 (1), Stats. for which he is eligible." Hauser, 82 Wis.2d at 55-57.
The Attorney General of the State of Wisconsin has also repeatedly confirmed the Department's method of computing good time under sec. 53.11. Assistant Attorney General Platz, generally viewed as a leading authority on Wisconsin criminal law, adhered to the Department's computations in computing good time in three attorney general opinions. See 29 Wis. Op. A.G. 135 (1940) (a prisoner with a 10 year sentence would be eligible for discharge at the end of 6 years, 3 months); 33 Wis. Op. A.G. 262, 264 (1944) (a prisoner with a 20 year sentence would be eligible for discharge at the end of 11 years and 3 months); 37 Wis. Op. A.G. 397, 398 (1948) (a prisoner with a 1 year sentence would be eligible for discharge at the end of 11 months).
Commentators have also repeated, without challenge, the Department's methodology. See, e.g., David D. Cook, Defense of Criminal Cases in Wisconsin secs. 7.29, 7.49-7.55 (1987); Robert D. Donohoo, Wisconsin Criminal Practice Manual sec. 3.11 (1979); 1974 Manual, Legal Assistance to Inmates and Mental Patients, sec. 6.02G(2) (Univ. of Wis. Law School 1974); 1979 Manual, Legal Assistance to Inmates and Mental Patients, sec. 3.06B.2 (Univ. of Wis. Law School 1979).
The language and legislative history of the statute, the Department of Corrections' long-standing, widely disseminated and widely relied upon interpretation of the statute, the Department's rulemaking process, and the repeated references in case law, attorney general's opinions, and commentary to the Department's method of computing good time all lead us to conclude that the court of appeals and this court should refrain from substituting their interpretation of this statute for the long-standing interpretation of the agency charged with its administration.
For the reasons set forth, we conclude that the legislature intended good time to be calculated according to the method used by the Department of Corrections. Accordingly we reverse the decision of the court of appeals. By the Court. — The decision of the court of appeals is reversed.