On November 8, 2010, Williamstown police officers executed a search warrant at the defendant’s dormitory room and seized a quantity of marijuana. The dormitory room was approximately 700 feet from the Williams College Children’s Center, an accredited preschool facility. The defendant was charged by criminal complaint in the District Court *552with possession of a class D substance (marijuana) with intent to distribute, in violation of G. L. c. 94C, § 32C (a), committing this violation within 1,000 feet of a preschool facility, in violation of G. L. c. 94C, § 32J, commonly known as a “school zone” violation.1
On August 2, 2012, the Governor signed into law St. 2012, c. 192, entitled “An Act relative to sentencing and improving law enforcement tools” (Crime Bill), which contained an emergency preamble that made it effective on enactment. Section 30 of the Crime Bill amended G. L. c. 94C, § 32J, by reducing the radius of the school zone from 1,000 feet to 300 feet. The defendant moved to dismiss the school zone violation, claiming that § 30 of the Crime Bill applies to all cases alleging a school zone violation that had not been adjudicated before August 2, 2012, and that his alleged violation occurred outside the amended school zone. The judge reported without decision the following question to the Appeals Court:
“Whether [St. 2012, c. 192, § 30], which reduces the radius of the Drug-Free School Zone from 1,000 feet to 300 feet, should be applied retroactively to an offense that occurred prior to the effective date of the amendment, but for which the Defendant had been charged but not adjudicated on the effective date of the amendment?”
We allowed the defendant’s application for direct appellate review. We answer “yes” to the reported question and hold that St. 2012, c. 192, § 30, applies to all cases alleging a school zone violation for which a guilty plea had not been accepted or conviction entered as of August 2, 2012, regardless of whether the alleged violation was committed before August 2, 2012.
Discussion. The temporal application of a penal statute is governed by the rule of statutory construction in G. L. c. 4, § 6, Second, which provides in pertinent part that the “repeal of a statute shall not affect any punishment, penalty or forfeiture incurred before the repeal takes effect, or any suit, prosecution or proceeding pending at the time of the repeal for an offence *553committed.” See Commonwealth v. Dotson, 462 Mass. 96, 99 (2012). Because we interpret an amendment of a penal statute to constitute an implicit repeal where “amended sections of a statute are inconsistent with the earlier provisions,” id. at 100, citing Nassar v. Commonwealth, 341 Mass. 584, 589 (1961), and deem a “punishment, penalty or forfeiture” to be “ ‘incurred,’ within the meaning of [G. L. c. 4,] § 6, Second, at the time the offence for which punishment is imposed is committed,” Dotson, supra, quoting Patrick v. Commissioner of Correction, 352 Mass. 666, 669 (1967), “a newly enacted [penal] statute is presumptively prospective.” Commonwealth v. Galvin, ante 286, 290 (2013). The consequence of this presumption is to “preserve, even after legislative change of a statute, the liability of an offender to punishment for an earlier act or omission made criminal by the statute repealed in whole or in part.” Dotson, supra at 99-100, quoting Nassar, supra.
This presumption of prospective application, however, is not absolute because the preamble to G. L. c. 4, § 6, declares, “In construing statutes the following rules shall be observed, unless their observance would involve a construction inconsistent with the manifest intent of the law-making body or repugnant to the context of the same statute . . . .” The question then is whether the prospective application of § 30 would be “inconsistent with the manifest intent of the law-making body or repugnant to the context of the same statute” (emphasis added). We note that, by separating these exceptions to the general presumption of prospective application with the word “or,” the Legislature intended that there be two exceptions, perhaps often related in fact, but separate and distinct in meaning. See Lobisser Bldg. Corp. v. Planning Bd. of Bellingham, 454 Mass. 123, 129 (2009) (fundamental to statutory constmction that word “or” is disjunctive unless context and main purpose of all words require otherwise), citing Bleich v. Maimonides Sch., 447 Mass. 38, 46-47 (2006), and Eastern Mass. St. Ry. v. Massachusetts Bay Transp. Auth., 350 Mass. 340, 343 (1966). See also Casa Loma, Inc. v. Alcoholic Beverages Control Comm’n, 377 Mass. 231, 234 (1979) (“It is a common tenet of statutory construction that, wherever possible, no provision of a legislative enactment *554should be treated as superfluous”). We have not before distinguished the different meanings of these two exceptions. We do so here.
1. “Inconsistent with the manifest intent of the law-making body. ” The presumption of prospective application is “inconsistent with the manifest intent of the law-making body” where there is “a clearly expressed intention” of the Legislature that the new statute apply retroactively. Nassar, 341 Mass. at 590. In ascertaining the intent of the Legislature, we look to “all [the statutory] words construed by the ordinary and approved usage of the language, considered in connection with the cause of its enactment, the mischief or imperfection to be remedied and the main object to be accomplished to the end that the purpose of its framers may be effectuated.” Galvin, supra at 290-291, quoting Hanlon v. Rollins, 286 Mass. 444, 447 (1934). To overcome the presumption of prospective application through this first exception, inferring that the Legislature probably intended retroactive application is not enough; that intent must be “clearly expressed.” See Dotson, 462 Mass. at 101 (“we see no clearly expressed intention by the Legislature to have the 2009 amendment to G. L. c. 272, § 53, applied retroactively”).
The Legislature may clearly express its intent through the words used in a statute or the inclusion of other retroactive provisions in the statute that would make prospective application of the provision at issue “anomalous, if not absurd.” Galvin, supra at 291. Thus, in Galvin, supra at 286-287, we concluded that the reduction in a mandatory minimum sentence required by § 14 of the Crime Bill for violations of G. L. c. 94C, § 32A (d), applied retroactively to a defendant who committed an offense prior to the effective date of the statute but whose conviction and sentencing did not occur until after that effective date. We noted that § 48 of the Crime Bill provided that those persons already serving a mandatory minimum sentence under § 32A (d) would be eligible for parole, probation, work release, and deductions in sentence for good conduct. Id. at 287. We concluded that the Legislature’s intent to make the reductions in mandatory minimum sentences retroactive under § 14 was manifest because, otherwise, the opportunity for a reduction in sentence under the Crime Bill would be provided to those who had already been sentenced for *555violations of § 32A (d), but not to “the limited class of persons who committed offenses before the amendments but were not convicted and sentenced until after the amendments’ effective date.” Id. at 291.
Here, the Legislature did not clearly express an intention that § 30 apply retroactively. There is nothing in the language of § 30 to reflect an intent that the reduced school zone radius apply either retroactively or prospectively; the section is silent as to its temporal application. Nor, in contrast with the Galvin case, are there other provisions in the Crime Bill, such as provisions applying the reduced school zone radius to those already sentenced for a school zone violation, that would make prospective application of § 30 “anomalous, if not absurd.” We also are not persuaded by the defendant’s argument that the inclusion of a preamble declaring the Crime Bill “an emergency law” whose “deferred operation . . . would tend to defeat its purpose,” suggests a clearly expressed intent that § 30 be applied retroactively. “The inclusion of an emergency preamble demonstrates only that the Legislature intended the statute to take effect without regard for the ninety-day waiting period otherwise provided by art. 48 of the Amendments to the Massachusetts Constitution.” Smith v. Massachusetts Bay Transp. Auth., 462 Mass. 370, 377 (2012). It does not suggest an intent to apply the Crime Bill retroactively. See Federal Nat’l Mtge. Ass’n v. Nunez, 460 Mass. 511, 521 (2011) (“While it is plain that the Legislature wished the new statute to take effect immediately to protect tenants in foreclosed properties from eviction, neither the emergency preamble nor the statutory text demonstrates a clear intent to apply the statute to properties already foreclosed on”).
2. “Repugnant to the context of the same statute. ” Our case law has yet to elaborate on the meaning of the phrase “repugnant to the context of the same statute,” but we can discern its meaning from the words themselves. Black’s Law Dictionary 1419 (9th ed. 2009) defines “repugnant” as “ [inconsistent or irreconcilable with; contrary or contradictory to,” and “repugnancy” as “[a]n inconsistency or contradiction between two or more parts of a legal instrument (such as a . . . statute).” Therefore, the presumption of prospective application is “repugnant to *556the context of the same statute” where it would be contrary to the purpose of the statute to delay the accomplishment of that purpose. While the phrase does not refer to the intent of the Legislature, and certainly does not require that the intent of the Legislature be made “manifest,” it does compel us to discern the legislative purpose of the statute at issue and determine whether prospective application would be inconsistent with that purpose.
Although framed as a separate crime, a school zone violation under G. L. c. 94C, § 32J, is effectively a sentencing enhancement for those who commit drug crimes involving more than mere possession within a school zone, because it requires imposition of a mandatory minimum term of imprisonment of two years to begin on the expiration of the sentence for the underlying drug crime.2 The sentencing enhancement does not require any additional wrongdoing by the defendant; it is enough that the drug violation occurred within a school zone, regardless of whether the defendant knew he was within a school zone. See Commonwealth v. Roucoulet, 413 Mass. 647, 650-651 (1992), quoting State v. Ivory, 124 N.J. 582, 593 (1991) (“After the elements of [the predicate] offense have been established, one need only take out the tape measure to see if [the school zone provision of § 32J] has been violated”).
The legislative purpose of § 32J when it was first enacted in 1989 was to protect school children from drug dealers by creating drug-free school zones. Roucoulet, supra at 652. But since its enactment, various studies have shown that the 1,000-foot radius was overbroad, and that its overbreadth has had an unfair impact on those living in urban communities. As one study summarized:
“Though the statute aims to protect children, its patterns of conviction indicate that it has more effectively created a two-tiered system of drug sentencing in Massachusetts. Because schools are more numerous in dense urban areas, most urban residents — including most of the state’s Black *557and Latino residents — face longer mandatory minimum sentences for drug offenses than the state’s rural residents, who are predominantly White.”
A Kajstura, P. Wagner, & W. Goldberg, The Geography of Punishment: How Huge Sentencing Enhancement Zones Harm Communities, Fail to Protect Children (2008). A study of Hampden County “found that residents of urban cities and towns are five times as likely to live in a sentencing enhancement zone as rural residents.” Id. Because African-Americans and Latinos comprised most of Hampden County’s urban population, they were more likely to live in school zones. Id. (finding majority — fifty-two per cent — of African-American and Latino residents in Hampden County lived in school zones compared to only twenty-nine per cent of white residents). At least in part because of the greater likelihood that African-American and Latino residents live in school zones, in 2011, seventy-three per cent of those convicted of school zone offenses in Massachusetts were racial or ethnic minorities even though they comprise less than one-quarter of the Massachusetts population and do not have a higher rate of illicit drug use. See Massachusetts Sentencing Commission, Survey of Sentencing Practices: FY 2011, at 85 (May 2012); United States Census Bureau, 2012 State & County QuickFacts, Massachusetts; Massachusetts Department of Public Health, Alcohol Use, Illicit Drug Use, and Gambling in Massachusetts, 2002, at 35-36 (July 2005). See generally J. Greene, K. Pranis, & J. Ziedenberg, Disparity by Design: How Drug-Free Zone Laws Impact Racial Disparity — and Fail to Protect Youth (2006). Moreover, because of the overbreadth of the 1,000-foot school zone radius, few school zone cases involved the sale or distribution of drugs to minors or minors’ participation in dmg-dealing activities. A study of drag dealing in three cities in Massachusetts — Fall River, New Bedford, and Springfield — found that seventy-one per cent of dmg-dealing incidents within school zones occurred when school was not in session. W.N. Brownsberger, S.E. Aromaa, C.N. Brownsberger, & S.C. Brownsberger, An Empirical Study of the School Zone Anti-Drug Law in Three Cities in Massachusetts, Journal of *558Drug Issues, at 936 (2004).3 Of the 340 drug-dealing incidents that occurred within a school zone, no more than four cases “involved charges of dealing to minors or using minors in sales.” Id. at 945 (Table 8) & 946.
The legislative history of § 30 demonstrates that the Legislature was aware of the school zone’s overbreadth and the disparate impact that resulted.- The chair of the House Judiciary Committee, Representative Eugene O’Flaherty, told the House chamber when it considered the conference committee report that included § 30:
“I hope you will understand what some of the urban districts have been dealing with. In Charlestown and Chelsea, you can’t stand anywhere in my district and not be in a school zone. If you are in Worthington, you can stand in Worthington and you are probably not going to be in a school zone. ... In urban areas all the individuals have this minimum mandatory sentence hanging over their head. . . . It has resulted in disparate sentencing.”
State House News Service, July 18, 2012, at 10-11. Similar arguments were made in the State Senate. During a debate regarding the radius of the school zone, Senator James Eldridge noted that the existing school zone law punished city dwellers more harshly. State House News Service, Nov. 10, 2011, at 8. Senator Stephen Brewer claimed that the entire city of Boston is a school zone. Id. at 9. Senator Daniel Wolf argued that the existing school zone law fostered the discriminatory incarceration of the urban population and of minorities. Id. Perhaps because the existing G. L. c. 94C, § 32J, was recognized to be overbroad, the focus of the debate on § 30 was not whether, but by how much, the school zone radius was to be narrowed. The Senate bill proposed that the radius be reduced to 500 feet, id. at 3, and the Governor proposed one hundred feet. State *559House News Service, Nov. 2, 2011, at 1. The conference committee settled on 300 feet.
Where the radius of the school zone was reduced from 1,000 feet to 300 feet at least in part because the broader radius was recognized to create an unfair disparate impact on those residing in urban areas and, consequently, on minority residents, and where the broader radius did not better protect school children from drug dealers, we conclude that it would be “repugnant to the context of [that] statute” to apply the § 30 amendment to § 32J prospectively and prolong the unfair disparate impact that the preamendment § 32J was having on urban and minority residents. We recognize that, generally, when the Legislature narrows the statutory definition of a crime or reduces the possible sentence for a crime, prospective application “may be, in the defendant’s view, an unfair consequence of G. L. c. 4, § 6, Second, but it does not rise to the level of repugnancy.” Dotson, 462 Mass. at 101. What distinguishes this amendment is that it was enacted to diminish the unfair disparate impact of the prior statute on urban and minority residents. Prospective application, therefore, affects more than the individuals charged with school zone violations; it affects all urban communities by subjecting their residents to a greater likelihood of a school zone sentencing enhancement than residents in suburban and rural communities.
In that regard, this case is similar to Dorsey v. United States, 132 S. Ct. 2321, 2326 (2012), in which the United States Supreme Court held that a provision in the Fair Sentencing Act, 124 Stat. 2372 (2010) (act), that reduced the “crack-to-powder cocaine disparity [in minimum mandatory drug sentences] from 100-to-l to 18-to-l” applied to offenders who committed a crack cocaine crime before the effective date of the statute but had not yet been sentenced. Title 1 U.S.C. § 109 (2012), the Federal counterpart to G. L. c. 4, § 6, Second, provides that the “repeal of any statute shall not have the effect to release or extinguish any penalty, forfeiture, or liability incurred under such statute, unless the repealing Act shall so expressly provide.” The Court noted that the United States Sentencing Commission had criticized Congress’s decision to set the crack-to-powder cocaine mandatory minimum ratio at 100-to-1, “because *560sentences embodying that ratio could not achieve the [Federal] Sentencing Reform Act’s ‘uniformity’ goal of treating like offenders alike, because they could not achieve the ‘proportionality’ goal of treating different offenders (e.g., major drug traffickers and low-level dealers) differently, and because the public had come to understand sentences embodying the 100-to-l ratio as reflecting unjustified race-based differences.” Dorsey, supra at 2328. The Court declared that the more lenient penalties in the act may be applied retroactively even if Congress did not expressly say so because Congress expressed its intention of retroactive application by “fair implication.” Id. at 2331-2332. The Court concluded that Congress intended the more lenient penalties for crack cocaine crimes to be applied retroactively because the Sentencing Reform Act, 98 Stat. 1987 (1984), requires sentencing courts to apply the guidelines in effect on the date the defendant is sentenced, provided they do not call for higher penalties, and applying the act’s more lenient mandatory minimum sentences prospectively “would create disparities of a kind that Congress enacted the Sentencing Reform Act and the Fair Sentencing Act to prevent.” Dorsey, supra at 2332-2333. Although the Federal interpretative statute differs from G. L. c. 4, § 6, Second, and the Supreme Court discerned the “fair implication” of congressional intent in the context of Federal statutes governing sentencing guidelines that have no Massachusetts counterpart, the result reached by the Supreme Court in the Dorsey case and by us in this case is essentially the same: where the Legislature chose to amend a sentencing statute at least in part to diminish its unfair disparate impact on a subset of our population, we will not prolong the unfairness by prospective application.
We recognize, as did the Supreme Court, that retroactive application will create its own set of disparities. See Dorsey, 132 S. Ct. at 2335 (“We also recognize that application of the new minimums to pre-[a]ct offenders sentenced after [the effective date of the statute] will create a new set of disparities”). With retroactive application, a person who committed a school zone violation before August 2, 2012, and pleaded guilty to that violation before that date will have his case adjudicated under the preamendment G. L. c. 94C, § 32J, while another person *561who committed the same school zone violation on the same date but whose case was still pending on August 2, 2012, will have his case adjudicated under the amended § 32J. We acknowledge this unfortunate disparity, but if the price of avoiding it is to prolong the unfair disparate impact on urban communities in matters of sentencing by declining to apply § 30 retroactively, we think the price is too high.
Conclusion. We answer “yes” to the reported question and hold that St. 2012, c. 192, § 30, applies to all cases alleging a school zone violation for which a guilty plea had not been accepted or conviction entered as of August 2, 2012, regardless of whether the alleged violation was committed before August 2, 2012. The case is remanded to the District Court for further proceedings consistent with this opinion.
So ordered.
The defendant was also charged in a separate complaint with possession of a class B substance (cocaine), in violation of G. L. c. 94C, § 34, but this complaint is not at issue on appeal.
With some exceptions, a person sentenced to a mandatory minimum sentence at a house of correction for a school zone violation is eligible for parole after serving one-half of the maximum term of the sentence. See G. L. c. 94C, § 32J.
In addition to narrowing the radius of school zones, St. 2012, c. 192 (Crime Bill) limited the time period to which the violation applied. Before passage of the Crime Bill, a designated narcotics crime within a school zone constituted a school zone violation regardless of the time of day in which it occurred. Under § 31 of the Crime Bill, a designated narcotics crime within a school zone constitutes a school zone violation only between the hours of 5:00 a.m. and midnight. St. 2012, c. 192, § 31.