(dissenting, with whom Cowin and Sosman, JJ., join). The court holds that the procedural protections governing the removal of an inmate from the general prison population and his confinement to a department segregation unit (DSU) apply to decisions regarding the housing of two-thirds of all of the prisoners in the State’s only maximum security prison, Massachusetts Correctional Institution, at Cedar Junction (Cedar Junction). These protections are contained in the department’s regulations at 103 Code Mass. Regs. §§ 421.00 (1993) (DSU regulations). Their application to the housing of the majority of prisoners at Cedar Junction is contrary to their purpose and the judicial and regulatory history leading to their enactment. Therefore, I dissent.
This litigation arises from the 1995 operational reconfigura*765tian of Cedar Junction into two separately functioning and controlled wings, a maximum wing (East) and a minimum wing (West).1 There is overwhelming support in the record for the conclusion that the reconfiguration was necessary to secure the safety of prisoners and staff alike, in the context of an unprecedented influx of violent gang members.
The plaintiff prisoners challenge the decision-making process leading to their assignment to housing in the maximum wing on due process grounds. In my view, decisions regarding the assignment of prisoners to housing within the general population of a prison are beyond the reach of the DSU regulations and not properly subject to a due process challenge in these cases. To the extent that the regimented life of a prisoner housed in the maximum wing of the prison may be too harsh, it can be challenged on grounds of cruel and unusual punishment. To the extent that a decision by the superintendent to house a prisoner in the maximum wing is arbitrary or based on improper considerations, it may be subject to an equal protection challenge. But in no event, on the record before the trial judge and this court, is there a basis in law or fact to enter summary judgment for the plaintiffs on their due process claim.
1. Background. Cedar Junction is the State’s only maximum security prison. It houses the most dangerous and violent prisoners within the Massachusetts prison population. For security purposes, prisoners who have engaged in serious acts of violence or other dangerous behavior while incarcerated at other Massachusetts prisons are transferred there.2 Because a paramount concern and responsibility of the Department of Correction (department) is ensuring the safety of the institution’s *766staff and inmates in an unimaginably dangerous environment, the department must adjust the operation of the institution to meet the challenges of the inmates sent into its custody. It is uncontested that these challenges became significantly greater in the 1990’s, with a substantial increase in the number of criminals entering the prison system with lengthy sentences to serve, and with the influx of large numbers of young and violent gang members, many of whom are now serving their sentences at Cedar Junction. As found by the Superior Court judge who considered and denied plaintiffs’ motion for injunctive relief in this case, “the recent infusion into the state prison of substantial numbers of members of warring gangs has presented the corrections system with unprecedented problems in controlling the institution and making it reasonably safe for staff and inmates.”
In the 1980’s, both the minimum and maximum wings at Cedar Junction were operated in a so-called “open campus” manner. However, in the early 1990’s, violent and organized attacks on correction officers and other inmates, combined with intelligence gathered from the prison population suggesting the planning of further, organized, large-scale acts of violence within the prison, necessitated a significant reassessment and reorganization of the institution. The institution needed to control the inmates, not vice versa.
In 1993, in the aftermath of some particularly violent incidents, Cedar Junction instituted a three-phase program, with fewer privileges and more restrictions in the first phase, and more privileges and fewer restrictions in phases II and m. The violence continued, and by 1995, the program had been déemed unsuccessful. In June, 1995, the superintendent announced a major operational change to the institution. The three-phase system was replaced by a two-phase system devised to take full advantage of the physical separation and the design differences between the two prison wings.3 Phase I was to be located in the maximum wing, and phase II, in the minimum wing. Phase I *767was characterized by higher restrictions and fewer privileges, and was to be operated under high security protocols; phase II was'characterized by fewer restrictions and greater privileges, and was to operate as an incentive and reward for positive behaviors by other Cedar Junction prisoners. The announced purpose of this change was the creation of a safer environment.
Following the implementation of the two-phase living system, inmates entering Cedar Junction for initial evaluation, inmates from other institutions reclassified to higher security and sent to Cedar Junction, and inmates identified as members of “security threat groups” (e.g., gang members) were housed in the maximum wing. Other prisoners were housed in the minimum wing, until and unless their conduct warranted (in the discretion of the superintendent) their relocation. Prisoners could be transferred from the maximum wing to the minimum wing on the basis of behavior and cell availability. Prisoners could also be transferred to other, lower security institutions on successful completion of a two-phase therapeutic program designed to break the cycle of gang involvement both in prison and after release.4
Although they are geographically within the same prison complex, the maximum and minimum wings are physically separate from each other and operate completely diiferently with regard to meals, recreational opportunities, and privileges. The restrictions on inmate movement in the maximum wing are substantially greater than those on inmates housed in the minimum wing and are similar in some respects to restrictions in the former DSU. The point of greatest similarity is the amount of time an inmate spends out of his cell. In the DSU, an inmate was permitted out of his cell one hour a day, five days a week. In the maximum wing, inmates are permitted out of their cells between one and one and one-half hours a day, seven days a *768week.5 Even at this point of similarity, there are significant differences. For example, if inmates chose to spend their out-of-cell time exercising, DSU inmates were released to an exercise cage (ten feet by thirty-five feet by ten feet) where they spent their time alone, while maximum wing inmates are released in groups of up to fifteen to a recreation yard measuring fifty feet by one hundred feet.
2. The DSU regulations. The regulations contained in 103 Code Mass. Regs. §§ 421.00, were enacted as the result of a series of orders of a single justice of this court in the case of Hoffer vs. Fair, No. SJ-85-0071 (Mar. 3, 1988). That case, commenced in 1985, dealt with the relationship between conditions under which the general prison population of Cedar Junction then functioned and the operation of the DSU as it then existed. It is reasonable to conclude from the records of both cases that the entire population of Cedar Junction had fewer restrictions and more privileges in 1985 than those currently afforded to any prisoner now housed there.
In 1985, prisoners who posed a danger to other prisoners, prison property, or the safe operation of the prison while living within the general population could be removed from that population and placed in a DSU where conditions were significantly more restrictive. The single justice concluded that the removal of such a prisoner should be subject to specific due process protections (which he found were required by the Massachusetts Constitution) principally because “[t]he nature and extent of the reduction of liberty from that experienced in the general population is . . . significant.” He consequently directed that the regulations now at issue be promulgated to govern the movement of a prisoner from the general population to a DSU. The department did not appeal from the ruling of the single justice, and the merits of that ruling are not now before us.
By 1995, the use of the DSU as a tool to control violence within Cedar Junction’s highly volatile population by isolating a few troublemakers (it housed only two inmates) from the general *769population was patently ineffective, and its use was eliminated.6 Efforts by the department to eliminate the corresponding DSU regulations, however, were subsequently enjoined by a single justice of this court, and they remain in effect today.7
3. Summary judgment. The motion judge ruled on the parties’ cross motions for summary judgment that the DSU regulations applied to the confinement of a prisoner in the maximum wing. He based this ruling on three necessary subsidiary findings: (1) the conditions of confinement in the maximum wing “are nearly identical to those which existed in the former DSU”; (2) the “general population” of Cedar Junction is not the population housed in the maximum wing; and (3) the commissioner’s reorganization and labeling of the maximum wing as the “general population” was “pretextual,” “aimed at avoiding compliance with [DSU] regulations.” See Longval v. Commissioner of Correction, 404 Mass. 325, 328-329 (1989) (“the department and the commissioner may not sidestep statutory and regulatory provisions stating the rights of an inmate as to his placement in a D.S.U. by assigning as a pretext another name to such a unit”).
In its opinion, the court affirms this portion of the judge’s summary judgment ruling. Because the last two of the judge’s three subsidiary findings are not supported by undisputed facts in the record, his ruling on summary judgment should be reversed.8
The second of the judge’s subsidiary findings is that the population in the maximum wing is not the “general population” of Cedar Junction. This finding is necessary to his ultimate ruling because the DSU regulations apply only to the removal of prisoners from the “general population” and their placement in conditions of confinement involving a significant reduction of liberty from that afforded the “general population.” If the *770population in the maximum wing is the “general population,” there is no removal and no significant reduction in liberty when a prisoner is assigned to that wing for housing. The judge bases his conclusion that the maximum wing is not the “general population” exclusively on his first subsidiary finding that the conditions of confinement in the maximum wing are nearly identical to the conditions of confinement in the former DSU. Therefore, he concludes that “the position taken by the Department [that the maximum wing is the general population] is not persuasive.”
The judge’s conclusion does not follow logically from his first subsidiary finding. His reasoning is both circular and inconsistent with the appropriate legal standard to be applied. The “general population” of a prison is not defined by the severity of the conditions of confinement. There is little question that the department could have uniformly imposed the restrictions now present in the maximum wing on the entire prison without violating the single justice’s judgment in Hoffer, and without having to comply with the DSU regulations. The general population is simply where the majority of prisoners normally lives, absent confinement in a specialized unit within the prison.9 Although the judge rejects the department’s contention that the maximum wing is the general population of Cedar Junction, he makes no explicit finding of where that “general population” resides. We assume that it must exist somewhere. If we leave the location of the “general population” undefined, it is impossible to conclude whether the DSU regulations apply to a transfer from that population to another population.
If the entire prison, both the maximum and minimum wings, constitute the “general population” as the court asserts, the DSU regulations are facially inapplicable to a prisoner’s place*771ment in or transfer to either wing.10 Similarly, if the maximum wing constitutes the general population, the DSU regulations are likewise inapplicable to a prisoner’s placement in or transfer to that wing. If the minimum wing constitutes the “general population,” the DSU regulations presumably would apply to a prisoner’s placement in or transfer to the maximum wing. But to conclude that the “general population” is the population residing in the minimum wing one would have to view the present operation of Cedar Junction through the “Looking Glass.”
Four hundred and eighteen of the 634 housing cells in Cedar Junction are located in the maximum wing and are subject to high security protocols. The much smaller minimum wing is operated differently to encourage and reward positive behavior by inmates. By creating different and less restrictive conditions within a discrete area of the general population of the prison as an incentive and reward for the positive behavior of prisoners, the department has not converted the prison into one in which those more beneficent conditions constitute the general conditions of its prison population. Nor have the department’s actions in creating a different living environment for a minority of prisoners converted the living area of two-thirds of the prison into something other than the “general population” of the prison. In sum, there is no factual support, and certainly no support based on uncontested facts, for the judge’s finding that the maximum wing is not part of the general population of Cedar Junction.
The judge’s third subsidiary finding, that the reorganization of the maximum wing of the prison and its denomination as the “general population” of Cedar Junction was a pretext to avoid compliance with the DSU regulations, is similarly flawed. A “pretext” is an “[ostensible reason or motive assigned or assumed as a color or cover for the real reason or motive; false appearance, pretense.” Black’s Law Dictionary 1187 (6th ed. 1990). To succeed on their motion for summary judgment, the *772plaintiffs had the burden of demonstrating that there was no genuine issue as to any material fact and that they were therefore entitled to a judgment as a matter of law. Mass. R. Civ. P. 56 (c), 365 Mass. 824 (1974). Whether the commissioner’s stated reasons for creating the current configuration of Cedar Junction were a pretext is a classic example of a disputed issue of material fact.11
Consistent with the requirements of Mass. R. Civ. P. 56 (e), 365 Mass. 824 (1974), the commissioner filed affidavits from the current12 and former13 commissioners of the department, and *773the assistant deputy commissioner,14 all of which were “made on personal knowledge, ... set forth such facts as would be admissible in evidence, and . . . show affirmatively that the affiant is competent to testify to the matters stated therein.” Mass. R. Civ. P. 56 (e). The affidavits describe the changing racial composition of the prisoner population, the sharp increase in individuals labeled as gang members, and the accompanying increase in violence in the prison system. These affidavits provide a factual explanation for the changes in the configuration of Cedar Junction that is entirely unrelated to avoiding compliance with the DSU regulations.
The plaintiffs do not dispute the facts set forth in these affidavits. At best, the plaintiffs suggest that there is a difference of opinion as to the appropriate response of prison officials to the problem of incarcerating young, violent gang members. However, opinions are not facts. And a difference of opinion is not the same as a dispute as to material facts.
In his affidavit, the plaintiffs’ expert provides his opinion, based on his experience as a warden of a maximum security prison in Ohio, regarding the appropriate treatment of inmates who are believed to be involved in gang activity. His review of inmate files led him to conclude that the Massachusetts prison system has “overuse[d]” segregation and “overreact[ed]” to gang identifications. This opinion does not satisfy the plaintiffs’ burden of demonstrating that there is no genuine issue of material fact for summary judgment and that they are entitled to *774judgment as a matter of law.15 Indeed, if this claim had gone to trial, as I believe it should have, the jury would have had the opportunity to assess the soundness and credibility of the plaintiffs’ expert’s opinion. See Leibovich v. Antonellis, 410 Mass. 568, 573 (1991). The expert’s testimony would not necessarily have been dispositive on this issue, and the jury would have been entitled to believe or disbelieve all, some, or part of the expert’s testimony. See Banaghan v. Dewey, 340 Mass. 73; 79 (1959); Dodge v. Sawyer, 288 Mass. 402, 408 (1934). In the face of conflicting affidavits, the plaintiffs’ expert’s affidavit does not provide a basis on which the judge properly could have concluded that the reorganization of the maximum wing of the prison and its denomination as the “general population” of Cedar Junction was a pretext to avoid compliance with the DSU regulations.16
The court’s opinion dismisses the content of the commissioner’s submissions that present the department’s reasons for the reorganization of Cedar Junction as “conclusory.”17 Ante at 754. I disagree on both points. The affidavits are not “conclusory,” but rather describe the facts and circumstances that led to the creation of the current configuration of the Cedar Junction population. It is not “conclusory” to state that in 1994, there were 463 correction officers assaulted by inmates in the Massachusetts prison system, or that between February, 1992, and *775February, 1995, the number of inmates in the Massachusetts prison system increased nearly ten per cent, from 9,970 to 10,835, while the number of correction officers decreased 3.5%, from 3,366 to 3,249. It is not “conclusory” to state that in 1995, 99% of the inmates at Cedar Junction were placed there because they had been violent and disruptive in other facilities. It is not “conclusory” to state that the ratio of inmates to correction officers at Cedar Junction in 1995 was greater than seven inmates to each correction officer at any given time. It is not “conclusory” to state that between March 1, 1995, and November 29, 1995, there were fifty incidents at Cedar Junction that required correction officers to use force against inmates. It is not “conclusory” to describe a July, 1993, confrontation between groups of prisoners at Cedar Junction that resulted in two prisoners being sent to a hospital in critical condition. And it is not “conclusory” to describe the attempted murder of a Cedar Junction correction officer in April, 1995, by a group of inmates, believed to be gang affiliated, who beat and repeatedly stabbed the officer, and the riot that ensued.
These were the facts that the commissioner presented to the judge to demonstrate that there is a disputed issue of material fact about the “real” reason the commissioner created the current configuration of Cedar Junction: whether the maximum and minimum wings were created in response to the increasingly violent and gang-oriented prison population, or whether they were created for the purpose of bypassing the due process requirements in the DSU regulations. Although the court’s opinion relies solely on the conditions of confinement as the basis for its conclusion that the DSU regulations are applicable to prisoners transferred to or placed in the maximum wing, Longval v. Commissioner of Correction, 404 Mass. 325, 328 (1989), requires more, as do the “pragmatic and flexible” requirements of procedural due process. Roe v. Attorney Gen., 434 Mass. 418, 427 (2001). On the record before him, the judge could not have found that there was no genuine dispute regarding this material fact. Summary judgment was therefore not appropriate.
4. Conclusion. The DSU regulations plainly do not apply to changes in the general conditions of confinement at any prison *776in the Commonwealth, or to the mere presence of severely restrictive conditions. They have no relevance to decisions about where (i.e., in which general population area) a prisoner will be housed, or to decisions to increase security requirements within different levels of confinement. They were promulgated to afford due process rights for those prisoners singled out for isolation from the general population in conditions significantly more restrictive than those experienced by that general population. The record does not support a finding on summary judgment that this is occurring at Cedar Junction. The DSU regulations apply to the current operation of Cedar Junction, if at all, only if prisoners were selected for further segregation from the maximum wing or the minimum wing populations, under conditions significantly more restrictive than those present in either.
For these reasons, I dissent.
The Massachusetts Correctional Institution, at Cedar Junction, is a maximum security prison. It consists of two wings, both of which house prisoners under maximum security conditions. The conditions of confinement in each wing vary, however. I refer to the West Wing as the “minimum” wing and the East Wing as the “maximum” wing as a way of describing the degree of confinement in each wing within the context of a maximum security prison.
The Department of Correction’s classification regulation, 103 Code Mass. Regs. § 420.09 (4)(a) (1995), requires a classification hearing whenever an inmate is transferred to a higher security correctional facility because of disciplinary or security issues. It is not contended in the cases before us that inmates transferred to Cedar Junction for disciplinary or other “security issues” have not received classification hearings with respect to those transfers.
Cedar Junction was designed and built with two separate wings. The maximum wing has eight cell blocks of forty-five inmates, organized on three tiers of fifteen cells each, with “grill” doors (bars), all facing an observation wall of one-way glass so that they can be closely monitored. It also contains a cell block for new prisoners. The minimum wing has three cell blocks of seventy-two inmates, organized on one floor (no tiers), with solid doors for *767more privacy. Each wing was also built with its own department segregation unit (DSU). Prior to 1995, there were thirty DSU cells in the maximum wing and sixty DSU cells in the minimum wing. Aside from the DSU cells, there are 418 cells in the maximum wing and 216 cells in the minimum wing.
As of May, 1998, there were 171 gang or security threat group members who successfully completed the entire program. One hundred fifty-six of these prisoners were reclassified and transferred to lower security prisons.
Although the Superior Court judge appears to have assumed that inmates in the maximum wing are released from their cells only five days a week, that assumption is contrary to the evidence in the record.
The former DSU cells are now used either to hold inmates awaiting disciplinary action or, in some instances, to hold inmates who are serving disciplinary sanctions as an overflow from the prison department disciplinary unit (DDU).
This order was also not appealed.
For these purposes, I assume that the judge’s finding that conditions of confinement in the former DSU and the maximum wing are nearly identical is adequately supported in the record.
The “general population” of a Massachusetts correctional facility is defined as “any housing area other than a special management unit, health service unit, departmental segregation unit, departmental disciplinary unit, or the departmental protective custody unit.” 103 Code Mass. Regs. § 423.06 (1995). This definition expressly does not apply to inmates housed in a “departmental segregation unit [or] a departmental disciplinary unit.”
The DSU regulations define a DSU as an area designated “for any inmate segregated” from the “general population.” 103 Code Mass. Regs. §§ 421.06, 421.09 (1993).
We are called on to deal with the concept of “pretext” in many other contexts. For example, in employment discrimination cases where a plaintiff must prove that an employer’s stated justification for an adverse action is not genuine, but rather is a pretext, we examine whether the employer’s proffered reason “has no reasonable support in the evidence” or is “wholly disbeliev[able].” See, e.g., Wheelock College v. Massachusetts Comm’n Against Discrimination, 371 Mass. 130, 138 (1976). Injury cases, where parties may not strike jurors based on factors such as race or gender, we examine whether the proffered explanation for striking a juror is “bona fide or a pretext” by considering “whether the challenge has a substantive basis or is impermissibly linked to race[, gender, or another factor].” See, e.g., Commonwealth v. Calderon, 431 Mass. 21, 26 (2000). In search and seizure cases, we look to see whether the reasons a police officer gives for conducting a search was a pretext concealing an investigatory motive. See, e.g., Commonwealth v. Matchett, 386 Mass. 492, 509-510 (1982).
Commissioner Michael Maloney attested in a 1998 affidavit that the department experienced “an unprecedented growth in the Massachusetts prisons of the number of volatile and frequently gang-affiliated incidents with younger inmates.” He explained that organized groups of inmates posed a significant threat to safety in the prison environment, and required the department to find a way effectively to manage, through incentives or otherwise, a violent prisoner population.
Former Commissioner Larry E. DuBois attested in a 1995 affidavit that, as of June, 1995: “the climate in the state prison system had become more dangerous, due no doubt in part to severe overcrowding, inmates serving longer sentences, increased racial friction and a growing number of volatile, often gang-affiliated, younger inmates.” He also states that: “In the past few years in Massachusetts, as law enforcement has secured the conviction of many gang leaders and members, there has been a corresponding unprecedented increase in gang membership in the Department’s prisons.” In support of this assertion, DuBois provides information about the size and names of different “security threat groups” (gangs and inmate organizations) in the Massachusetts prison system. At that time, “[t]he ever increasing danger of the situation is reflected in the fact that, during 1994 alone, 463 correctional officers were assaulted by inmates.” DuBois also described the serious injuries that inmates inflicted on three guards, which led to the lockdown in 1995.
Assistant Deputy Commissioner Ronald Duval attested in a 1998 affidavit that: “During my twenty-five (25) years at MCI-Cedar Junction I have personally observed dozens of changes in the operational parameters of the prison. Every change was for the sole purpose of effectively managing the prisoner population in the Commonwealth’s only maximum security prison. The extent to which the operational parameters of the [maximum] and [minium] wings have continually been adjusted, altered, and changed throughout the past 25 years, and more frequently in the past 10 years, is a direct result of corrections officials having to manage an increasingly violent prisoner population throughout the Department with limited resources.” He noted that the changes to the maximum and minium wings “would drastically reduce violent acts among the prisoners.”
In its opinion, the court suggests that the defendants should have provided additional evidence to support their assertions about the increase in gang members and violence in Massachusetts prisons. Ante at 758-759. However, the commissioner does not bear the burden of proof as to the plaintiffs’ motion for summary judgment; that burden is borne by the plaintiffs alone. The commissioner is not required to set forth his entire defense to the plaintiffs’ claims to defeat a motion for summary judgment. The affidavits that the commissioner provided were sufficient to create a dispute concerning a material fact, which is all that is required to defeat such a motion.
The court also focuses on the conclusion in the expert’s affidavit that less than fifteen per cent of a random sample of 110 prisoners in the maximum wing had received a follow-up classification review within twelve months of their previous classification review. This reference is not relevant to the claim regarding the applicability of the DSU regulations to the maximum wing because classification hearings are governed by an entire separate regulatory scheme. See 103 Code Mass. Regs. §§ 420.00.
The court does not challenge the credibility of the affiants. Of course, credibility is not an issue to be resolved on summary judgment. See Attorney Gen. v. Brown, 400 Mass. 826, 832 (1987).