dissenting:
My colleagues rightly point out that the “collective knowledge” doctrine is well-established in this jurisdiction. But the majority holding in this case stretches that doctrine to the breaking point in concluding that a Terry stop may be justified by knowledge that is never communicated in any form whatsoever to the officers who make the stop. This holding disregards a prior opinion of this *1135court and goes well beyond established precedent in this or any other jurisdiction. Accordingly, while I have no quarrel with parts I — III of the majority opinion, I must respectfully dissent from part IV and, as a consequence, do not need to address part V.
In its usual form, the collective knowledge doctrine, also known as the “fellow officer rule,” stands for the common sense principle that an officer who seizes a suspect need not have personal knowledge of all the facts and circumstances that justify the seizure, as long as the officer is responding to a directive or other communication that was itself based upon facts sufficient to warrant this action.1 This last proviso is important; typically the courts have refused to extend the doctrine to situations where the seizing officer is not acting either on a validly issued directive or on some other form of communication that, combined with his or her own observations, justifies the stop. Thus, in Whiteley v. Warden, Wyoming State Penitentiary, 401 U.S. 560, 91 S.Ct. 1031, 28 L.Ed.2d 306 (1971), the Supreme Court freed a state prisoner who had been arrested pursuant to a bulletin that had been issued on the basis of an invalid warrant. In concluding that the affidavit filed by the sheriff to obtain the warrant was defective because it was barren of any factual detail, id. at 564-65, 91 S.Ct. at 1035, the Court also noted that the defective affidavit could not be rehabilitated by additional information allegedly known to the sheriff when he filed the affidavit but never communicated to the magistrate issuing the warrant, id. at 565 n. 8, 91 S.Ct. at 1035 n. 8, nor could the arrest be validated by the fact that the officers were relying in good faith on the bulletin, id. at 568, 91 S.Ct. at 1037. By comparison, in United States v. Hensley, 469 U.S. 221, 105 S.Ct. 675, 83 L.Ed.2d 604 (1985), the Court found that a Terry stop based on a flyer issued by police in another state was valid because the issuing police had reasonable grounds for making such a stop themselves.
Assuming the police make a Terry stop in objective reliance on a flyer or bulletin, we hold that the evidence uncovered in the course of the stop is admissible if the police who issued the flyer or bulletin possessed a reasonable suspicion justifying a stop and if the stop that in fact occurred was not significantly more intrusive than would have been permitted the issuing department.
Id. 469 U.S. at 233, 105 S.Ct. at 682 (emphasis in original,' citation omitted).
Accordingly, while the seizing officers need not know all the facts and circumstances justifying the seizure, there must be a connection — established by a valid directive or other communication — between these justifying facts and the resulting seizure. We observed in Haywood v. United States, 584 A..2d 552, 556-57 (D.C.1990) (citations omitted, emphasis in original):
While the collective knowledge of the police can give rise to a valid arrest, this is so only if the arresting officer acts in response to a broadcast or other directive which is based on the collective information.... An arresting officer need not have firsthand knowledge of the facts giving rise to probable cause provided that he or she is acting at the suggestion of someone who does.... In cases ... where probable cause for arrest is predicated in part on the personal observations of the arresting officer, the court may not rely on facts which were available to other officers at the scene unless that information was communicated to the arresting officer.2
In other words, all the information imputed to the seizing officer must normally have been communicated to that officer in some form, however abbreviated, in order to be *1136considered by the court in determining whether the stop was adequately justified.3
This is not to say that any single officer must know individually all of the elements necessary to warrant a stop. See Smith v. United States, 123 U.S.App.D.C. 202, 204, 358 F.2d 833, 835 (1966) (citing Ker v. California, 374 U.S. 23, 83 S.Ct. 1623, 10 L.Ed.2d 726 (1963)), cert. denied, 386 U.S. 1008, 87 S.Ct. 1350, 18 L.Ed.2d 448 (1967). Moreover, the seizing officers may also supplement any directive they receive with other information that they subsequently discover on their own. See Smith, 123 U.S.App.D.C. at 205, 358 F.2d at 836. My point is simply that the government cannot justify a seizure by relying on facts that never played any part in the chain of reasoning that led to the seizing officers’ decision to make the stop.
Both Smith and Woodward v. State, 668 S.W.2d 337 (Tex.Crim.App.1982) (en banc), cert. denied, 469 U.S. 1181, 105 S.Ct. 939, 83 L.Ed.2d 952 (1985), which my colleagues cite in support of their conclusion, fit within this understanding of the collective knowledge doctrine. In both cases, the knowledge that was imputed to the arresting officer had been communicated to him in some form. In Smith, the informant’s tip was relayed to the arresting officer by another officer, albeit without specific information about the informant’s reliability. In Woodward, the arresting officer knew that there had been a statewide bulletin issued by the Austin police to stop the defendant for questioning in connec*1137tion with a homicide and that the defendant had recently been sighted 90 miles away from the scene of the crime. In each ease the arresting officer then added to this information further relevant knowledge of his own. Thus, the reviewing court was entitled to consider both the information underlying the initial directive and the additional facts later known to the arresting officer in considering whether there was an adequate ground for the stop in question.
In this case, by contrast, the majority concludes that the stop of appellant was justified by critical information that played no part in the lookout broadcast and was never made known in any form to the seizing officers — or to anyone else, for that matter — before appellant was stopped. My colleagues suggest that this information eventually became part of the collective knowledge underlying the dispatcher’s initially improper lookout instruction, and should be deemed a part of that collective knowledge before appellant was stopped, for one reason: without telling a soul, Detective Reed consciously decided, after learning the additional information, that any rebroadcast of the original (defective) lookout was unnecessary. But silence cannot be an adequate basis for justifying a search or seizure; there must be an affirmative act or acts communicating sufficient relevant knowledge, either directly or indirectly — for example, through a dispatcher or a flyer — to the seizing officers.4 There may, of course, be circumstances where an initial tip or directive, inadequate in itself to support the actual seizure, is later supplemented by additional information, discovered by or relayed to the arresting officer, that justifies the stop, as in Smith and Woodward. But I know of no case holding, as the majority does here, that an invalid directive may be retroactively rehabilitated by new information that is never communicated in any form, through a dispatcher or otherwise, to the officers who actually make the stop.5
My colleagues claim that requiring an officer in Detective Reed’s position to initiate a new lookout6 would not provide any additional protection to citizens from unlawful police conduct. I emphatically disagree. The position adopted by the majority would allow the police, without sanction, to issue and act on premature seizure orders in the expectation that, by the time of the seizure, someone somewhere in the police system will have adequate information justifying detention. Taken to its logical conclusion, this view would permit random sweeps of allegedly “high crime” neighborhoods to check identities and pick up persons who are wanted on outstanding warrants.7 It is because of this *1138danger that jurisdictions adopting the collective knowledge doctrine have carefully confined it to situations where the seizing officers are acting on the basis of a valid seizure order and/or other information that they discover or that is communicated to them in some form. As the Colorado Supreme Court has warned, “The fellow officer rule ... is not a means of creating probable cause by using post hoc combinations of information available to the police.” Hazelhurst, 662 P.2d at 1087.
I vote to reverse and remand with an order that appellant’s confession, tainted by his unlawful arrest, must be suppressed.
. See, e.g., Williams v. United States, 113 U.S.App.D.C. 371, 372, 308 F.2d 326, 327 (1962) (“in a large metropolitan police establishment the collective knowledge of the organization as a whole can be imputed to an individual officer when he [or she] is requested or authorized by superiors or associates to make an arrest”).
. It is noteworthy that in making these observations, the Haywood court cited Smith v. United States, 123 U.S.App.D.C. 202, 358 F.2d 833 (1966), cert. denied, 386 U.S. 1008, 87 S.Ct. 1350, 18 L.Ed.2d 448 (1967), upon which the majority relies so heavily here.
. See United States v. Edwards, 885 F.2d 377, 382 (7th Cir.1989) ("A supervising officer’s knowledge about a defendant cannot be relied upon to provide probable cause for his arrest where there is no evidence that such knowledge was communicated to the agents on the scene who actually made or ordered the defendant’s arrest.”); United States v. Woods, 544 F.2d 242, 259 (6th Cir.1976) (superior officer’s knowledge cannot be considered in determining probable cause, where there is no evidence that his order was basis of defendants’ arrests), cert. denied, 429 U.S. 1062, 97 S.Ct. 787, 50 L.Ed.2d 778, 430 U.S. 969, 97 S.Ct. 1652, 52 L.Ed.2d 361, (1977); People v. Hazelburst, 662 P.2d 1081, 1087 (Colo.1983) (fellow officer rule did not supply basis for detention of defendant, where arresting officer had been given no information implicating defendant, even though such information was known to investigating officer); State v. Cooley, 457 A.2d 352, 355 (Del.1983) ("To say in the abstract that probable cause is to be evaluated on the basis of the collective information of the police ignores the underlying assumption — and factual reality — that there is some communication between those officers, who do know facts amounting to probable cause, and those who do not.”); State v. Crowder, 1 Haw.App. 60, 613 P.2d 909, 915 (1980) (imputation of collective knowledge not available where there was no evidence that any officers communicated with each other or that arresting officer was acting on directive from fellow officer); Salter v. State, 163 Ind.App. 35, 321 N.E.2d 760, 762 (1975) (one officer’s knowledge of probable cause could not justify arrest by another officer, absent any evidence of communication between them); State v. Martin, 232 Neb. 385, 440 N.W.2d 676, 682 (1989) (fact that there was a warrant outstanding for defendant’s arrest does not make the arrest lawful, where this fact was not known to arresting officers and was not the basis for actual arrest); State v. Mickelson, 18 Or.App. 647, 526 P.2d 583, 584 (1974) (Police officers cannot “search on the hope that the total knowledge of all those officers involved in a case will later be found to constitute probable cause if the search is challenged.... Cases in which courts have adhered to [the collective information] principle are those where the arresting officer acted with an awareness or reasonable belief that fellow officers have information sufficient to constitute probable cause.... [S]omewhere in joint police action there must be a nexus between the probable cause and the invasion of privacy, between the justification and the act.”) (citations omitted); Commonwealth v. Gambit, 274 Pa.Super. 571, 418 A.2d 554, 557 (1980) ("In no instance has an arrest been justified where the information providing probable cause was known only to officers unconnected with the arrest. What is missing here is some communication or connection between [the arresting officer] and the officers who heard the ... broadcast”), aff'd, 501 Pa. 453, 462 A.2d 211 (1983).
In a very few cases, some courts have imputed knowledge known by one officer to a partner working as part of the same arrest team "in a close time-space proximity to the questioned arrest.” 2 Wayne R. LaFave, Search & Seizure § 3.5(c) at 17 (2d. ed. 1987). See United States v. Ragsdale, 470 F.2d 24 (5th Cir.1972) (knowledge of one officer could be imputed to partner where both involved in automobile stop, and first officer told second officer of gun in auto, but partner apparently did not hear this remark); Commonwealth v. Wooden, 13 Mass.App.Ct. 417, 433 N.E.2d 1234 (1982) (imputing knowledge of one officer to another, where both were working in concert and within arm’s reach of each other and of suspects). This exception — which in any case appears to conflict squarely with the language quoted in the text above from Haywood — • does not apply to the facts before us, given that Detective Reed was not a member of the team that stopped appellant and was far away from the scene of the stop.
. See cases cited supra at note 3.
. Indeed, in the one case with circumstances most closely approximating those before us, albeit in a jurisdiction that has not accepted the collective knowledge doctrine, the court rejected this proposition. See People v. Ford, 150 Cal. App.3d 687, 198 Cal.Rptr. 80, 88 (1984). Compare People v. Beard, 35 Ill.App.3d 725, 342 N.E.2d 343, 348 (1976) (information not specifically known by arresting officer could be considered in finding probable cause where this information was communicated to police headquarters and headquarters then reconfirmed earlier bulletin to stop and investigate).
. To make my position clear, I believe that it would not be sufficient for an officer in Detective Reed’s position merely to communicate the new information to the dispatcher. Only a new lookout broadcast, based on the new information, would satisfy the requirement that there be a connection between the information justifying the seizure and the seizure itself.
. I do not believe that this concern is unwarranted, as my colleagues suggest. The majority opinion relies on an extraordinarily broad formulation of the collective knowledge doctrine and then disregards, without adequate justification, a basic and widely recognized limitation on the application of that doctrine to situations such as the one in this case. In response to this dissent, my colleagues appear to acknowledge that the collective knowledge doctrine does not extend to the entire police force. They intimate, moreover, that particular knowledge may be added to the justification for a seizure only when that knowledge has been obtained by a member of the investigating team and when the delay between the improperly issued seizure directive and the seizure itself is relatively short. But such standards are, I fear, subject to considerable manipulation. Moreover, the position taken by my colleagues, instead of promoting a proper regard for citizens’ Fourth Amendment rights, tacitly approves sloppy police practice by justifying seizures on the basis of directives that admittedly were improperly issued. I see no reason to encourage such a practice. When we are justifying a seizure based, necessarily, on collective knowledge, I believe it is critical that we hold the police to an easily understood, bright-line definí*1138tion rather than permit an exception — as in this case — based on a single officer’s judgment that a shortcut is permissible. See Ford, supra note 5, 198 Cal.Rptr. at 88 ("no societal or legitimate police purpose exists for encouraging the issuance of a 'be on the lookout for’ memorandum in the first place without probable cause”).