concurring in part and dissenting in part.
I agree that the evidence is insufficient to convict James Andrews of marijuana possession under a theory of constructive possession, or as an aider and abettor. Because there is no meaningful distinction in the nature or quantity of evidence presented against Tawanna Bolden that crosses the threshold of reasonable doubt, I conclude that the evidence is insufficient to convict her under either theory of liability as well. Thus, I would reverse Bolden’s conviction of marijuana possession, and dissent from that part of the Per Curiam opinion affirming her conviction.
To frame my analysis, I begin by including some facts that are not recited in the Per Curiam opinion. Patricia Baker, the friend whose testimony was credited by the trial court, testified that Bolden was seriously ill with brain cancer and was “always sick” with symptoms that included fever, feeling cold, vomiting, and weakness. As a result, according to Baker, Bolden “sleeps all the time” and doesn’t “have any energy.” Baker corroborated Bolden’s testimony that on the day in question, before Baker and Andrews went out of the house leaving the front door unlocked, Bol-den had come downstairs briefly to prepare breakfast for her children, and had *538returned to her second floor bedroom to sleep, as was her custom. Baker also testified that when she and Andrews returned to the house one to two hours later, she had looked into the second floor bedroom and seen Bolden asleep, curled up under the covers. It was only a few minutes later that the police officers broke down the back door of the house, surprising the men who had been packaging marijuana in the downstairs kitchen, and rushed up the steps to catch them before they could hide or attempt escape. Bolden testified that she was awoken by the screams of her young children who, scared by the police onslaught, had come running up the stairs to where Bolden was sleeping on the second floor. Baker, who was in the second floor bathroom when the police arrived, corroborated this aspect of Bolden’s testimony as well, saying that as she came out from the bathroom, she saw Bolden emerge sleepily from her bedroom into the hall, looking scared and calling for her young son. No one testified to the contrary.
Notwithstanding Bolden’s and Baker’s testimony, my colleagues conclude that “the magnitude and duration of the drug activity taking place in the house while Bolden was present as the lessee and main occupant” were sufficient to enable the trial court to find that Bolden knew about the marijuana in the house and had facilitated the use of her house for the unlawful activity, thus becoming criminally liable as an aider and abettor. See ante at 535. I do not question that the police’s observation of a cloud of marijuana smoke and drug activity upon their arrival would, in the usual case, provide a factual basis from which the finder of fact could reasonably infer that an occupant of the house would similarly have been aware of the marijuana smoke. But that inference is not reasonable on the facts of this case. Bolden was asleep in a second floor bedroom when the police arrived. Even though there was, at most, a two-hour lapse between the time when Baker saw Bolden go upstairs to sleep and Baker’s return when she saw Bolden still asleep in her room — two hours during which it is theoretically possible that Bolden could have awoken and become aware of the marijuana smoke— there is no evidence in the record as to when the men entered the house during that period or as to when the marijuana smoke would have permeated the house so that it would have been noticeable on the second floor. Nor is there any evidence that Bolden in fact was awake at any time during those two hours. Baker’s testimony strongly implies that she was not. Any inference, therefore, that Bolden had knowledge of the marijuana smoke is sheer speculation.
Knowledge of criminal activity is a prerequisite, but by itself is not enough; aiding and abetting requires some affirmative assistance to commission of the crime. Usually assistance is shown by some action, even if slight, that abets in commission of the crime. To premise criminal liability on in action, as the government proposes here, there must be circumstances which in context support an inference that by failing to act a person associated with the venture and participated in it as in something that he or she wished to bring about. See Settles v. United States, 522 A.2d 348, 357 (D.C.1987) (quoting Nye & Nissen v. United States, 336 U.S. 613, 619, 69 S.Ct. 766, 93 L.Ed. 919 (1949)); cf. Montgomery v. United States, 384 A.2d 655, 659 (D.C.1978) (quoting United States v. Garguilo, 310 F.2d 249, 253 (2d Cir.1962) (“There may even be instances where the mere presence of a defendant ... will permit a jury to be convinced ... that the defendant sought by his action to make it [the crime] succeed — for example, the attendance of a 250-pound bruiser at a *539shakedown as companion to the extortionist .... ”) (internal quotation omitted)). Failure to act when action would be expected could make innocent presence less likely. See Settles, 522 A.2d at 357 (“[M]ere presence at the scene of the crime, without more, is generally insufficient to prove involvement in the crime, but it will be deemed enough ‘if it is intended to [aid] and does aid the primary actors.’ ”) (alteration in original) (citations omitted).
Turning to the facts of this case, a fact-finder could begin by assuming that a person with legal authority or responsibility would take steps to assert control over property and to keep her children safe from harm. But such assumption needs to be tested against the facts in evidence because a person’s inaction could be the result of something other than the wish to assist the commission of crime. Fear, a sense of loyalty to family or friends, and physical or mental incapacity, to name a few, could also explain why a person would not try to put a stop to illegal activity. The inference that inaction equates to acquiescence and assistance is particularly strained here, where Bolden did not act with consciousness of guilt when the police arrived and her serious illness and sleeping habits call into question her ability to notice what is going on in other areas of the house and take quick action in response. Although the circumstances were undoubtedly suspicious, suspicion alone does not support a finding of guilt without probative evidence that Bolden had the requisite intent. See Greer v. United States, 600 A.2d 1086, 1087-88 (D.C.1991) (discounting evidentiary value of suspicious circumstances). To conclude that Bolden intended to aid and abet on this record is not a reasonable inference from the evidence, but requires the factfinder to cross into impermissible speculation.
This is not a case like United States v. Jenkins, 289 U.S.App. D.C. 83, 928 F.2d 1175 (1991), where the court found the evidence “just barely” sufficient to convict a homeowner of conspiracy to possess cocaine relying on the “natural inference ... that those who live in a house know what is going on inside, particularly in the common areas.” Id. at 87, 928 F.2d at 1179. Present in Jenkins — but not the case here — was drug activity carried out not only in the common areas of the house, but by persons who lived in the house and, the factfinder could infer, “routinely” engaged in drug dealing on the premises in the view of the other residents. Id. Moreover, the homeowner was “closely associated” with one of the drug dealers. Id. at 86, 928 F.2d at 1178. Of particular importance to the court in Jenkins was .38 caliber ammunition found in the bedroom of the homeowner which, coupled with expert evidence that people who have drugs usually also have firearms, gave further weight to the homeowner’s complicity in the drug activity. See id. at 88, 928 F.2d at 1180. Under those circumstances, the inference that the homeowner is aware of drug dealing in her house and has facilitated use of the premises can reasonably be sustained. In this case, on the other hand, all we have is drug activity of, at most, two hours duration, carried out by persons who do not live in the house, and no evidence of any contraband — neither drugs nor weapons — associated with Bolden’s immediate living space.1 Nor is this case like Greer, where the tenant personally opened the door for the drug dealers and was found *540by the police in their company, attempting to hide in a closet full of drug paraphernalia while the drug dealers were dumping drugs out the window. See Greer, 600 A.2d at 1087. Because the tenant in Greer obviously had knowledge of the drug activity in her home, her action in allowing the drug dealers into the apartment and staying with them fully supported an inference that she facilitated their activity. There is no comparable evidence in this case that Bolden knew of the marijuana or made her home available to drug dealers. Here the fact-finder must have a reasonable doubt about what Bolden knew and why she acted (or failed to act) as she did. See Rivas v. United States, 783 A.2d 125, 134 (D.C.2001) (en banc) (stressing this court’s obligation to critically scrutinize the evidence to ensure that a rational fact-finder could find guilt beyond a reasonable doubt).
Beyond my conclusion that the evidence is insufficient, I am troubled by the majority’s approach in deciding to affirm the conviction. The trial court’s reasoning in finding Andrews and Bolden guilty was brief and to the point: The trial judge found Andrews guilty because he was a “lawful occupant” of the house and had “the power to admit others to the premises and to exercise control over the premises [] and its use.” As to Bolden, the trial judge noted that she too was a “lawful occupant,” that her name was on the lease documents and that she “had the authority to control access to, and use of, the premises.” The trial court concluded that it was “highly improbable” that five men “would enter Ms. Bolden’s house ... in the daylight hours, to smoke and package ... large amounts of marijuana ... without [her] knowledge and consent.”2
Thus, the record is clear that, as with Andrews, the trial court relied on Bolden’s status as lessee and lawful occupant with the authority to control the premises to infer acquiescence and find guilt. My colleagues recognize that the trial judge’s stark reasoning was incorrect and did not provide legal support for Andrews’s guilt, but conclude it is sufficient to sustain Bol-den’s guilty verdict. But what is the difference in the eyidence presented against them? Andrews and Bolden both resided in the house and both, as the trial judge found, had authority to control the premises. There is the one difference, emphasized by my colleagues: Bolden was the lessee, whereas Andrews was a roomer. But legal status, without more, is insufficient to prove guilt. If mere presence (ie., knowledge of the commission of crime in one’s presence) at the scene of a crime is insufficient without “conduct which designedly encourages or facilitates a crime” in order to support an inference of “guilty participation as an aider and abettor,” Garrett v. United States, 642 A.2d 1312, 1317 (D.C.1994), bare status must be less so as it does not even imply knowledge of the commission of crime in one’s presence. To hold otherwise is to make a homeowner or lessee strictly liable criminally for the illegal activities of others conducted on the premises over which the home owner or lessee has authority.
It also is evident from the record of this bench trial that the trial judge did not make the credibility determinations and fact-findings that would be necessary to support the alternative conclusion proposed by my colleagues, that Bolden became aware of the marijuana smoke with time to act such that her inaction would imply acquiescence. In doing so, they en*541gage in appellate fact-finding and overstep our appellate function of reviewing trial court action. A trial court need not make specific fact-findings and in the absence of findings we must affirm if the evidence of record is sufficient to support the result. That does not mean, however, that on appellate review we can ignore fact-findings and credibility determinations that were made by the trial judge. The trial judge credited Baker’s testimony that it was she and Andrews who had left the door unlocked when they left the house, even though Bolden usually kept it locked, and that they had returned home only five minutes before the police arrived, at which time Baker saw a cloud of smoke that smelled like marijuana. The trial court made no finding of fact as to whether Bolden became aware that there was marijuana in her house from the cloud of smoke that Baker and the police discovered upon their arrival, nor as to when she would have noticed it so that a reasonable inference could be drawn that, knowing of the presence of marijuana in her house, she acquiesced in the drug activity being carried out by others. To the extent that Baker’s credited testimony was relevant to these questions, however, her testimony that Bolden went to sleep before Baker left and was sleeping when she returned supports the opposite inference: that Bol-den was unaware of, and did not acquiesce in, the drug activity. Therefore, the inference on which my colleagues rely has no factual predicate in the evidence heard and credited by the trial judge to affirm Bol-den’s conviction.
Therefore, I dissent because the reasoning provided by the judge is incorrect as a matter of law, and an alternative rationale is not supported by the evidence.
. The government’s theory was that the marijuana was brought to the house by the drug dealers that day and that the marijuana found on the second floor had been dropped or hurriedly stashed away in a trash can and behind a bookcase as the men in the kitchen ran upstairs when the police broke in through the door.
. The trial court’s conclusion is undermined by the testimony it credited that Bolden usually kept the door locked and that it was Baker and Andrews who left it unlocked when they left the house that morning.