This sentencing appeal provides an opportunity to clarify the operative standards for identifying relevant conduct under U.S.S.G. Sec. 1B1.3(a)(1)(B) (Nov.1993).1 We seize the opportunity and, in the end, affirm the sentence imposed below.

For many years, defendant-appellant Evangelist LaCroix earned his livelihood as a building subcontractor in southern New Hampshire. He became acquainted with the brothers Zsofka, Matthew and Lazlos, who, through entities known as ZLM Realty and 101 Realty (the Zsofka entities), planned to develop a sizable single-family residential real estate complex known as "Sunview II." In late 1985, appellant and Matthew Zsofka (Zsofka), together with Zsofka's construction foreman, John Lee, formed a corporation, Alpha Construction Company, to serve as the general contractor for Sunview II. Appellant became Alpha's president, though by all accounts Zsofka retained ultimate control.

Construction and sales proceeded apace until the summer of 1987, when demand began to slacken. Alpha responded to adversity by retaining a marketing agent, Horns of New Hampshire (HNH), a firm headed by Richard Horn. Zsofka and Horn masterminded an illegal scheme that enabled their companies to market and sell roughly 90 homes over the following two years.

The conspirators' plan was seductively simple: they secretly gave money, secured by a late-filed second mortgage, to any would-be homeowner who lacked the wherewithal for the minimum down payment required by the prospective purchase-money mortgage lender (usually the Dime Savings Bank).

Appellant personally handled 31 closings at which he falsely represented, both orally and in writing, that no undisclosed financing arrangements existed. Appellant knew these statements to be apocryphal when made. The other 60-odd closings were handled in much the same fashion by one or the other of appellant's coconspirators. The transactions were structured in such a way that, on paper, Alpha conveyed the houses, but not the land, to the buyers. The company received in excess of $37,000 at every closing. These proceeds enabled Alpha, among other things, to assist the Zsofka entities in funding the clandestine second mortgages.

After Zsofka and Horn hatched the plot, appellant attended weekly staff meetings at which all the closings, including those handled by others, were discussed and approved. At no fewer than three of these meetings Zsofka preached to those present, appellant among them, about the importance of keeping all secondary financing hidden from the first mortgagees. Zsofka also gave instructions on how best to accomplish this furtive feat.

During the under-three-year period when the scheme was velivolant, appellant drew a total of approximately $385,000 in salary from Alpha. In sum, as a part-owner and salaried officer of Alpha, appellant participated in, or was present at the discussion of, every transaction, profited at least indirectly from each sale, and stood to gain more money later (when and if the buyers repaid the second mortgages).

Over time, many of the borrowers proved unable to pay the first mortgages, resulting in widespread foreclosures at a net cost to the Dime Savings Bank in excess of $2,800,000. Losses of this magnitude are seldom unremarked. In 1992, a federal grand jury returned a 102-count indictment against the three Alpha principals and four persons associated with HNH. The indictment charged appellant with conspiracy to defraud a federally insured financial institution in violation of 18 U.S.C. Sec. 371, and with various substantive offenses, including 12 counts of bank fraud, 18 U.S.C. Sec. 1344, and 12 counts of making false statements to a federally insured financial institution, 18 U.S.C. Sec. 1014. After a 17-day trial, the jury announced its inability to reach agreement on the 24 counts charging appellant with the commission of substantive offenses,2 but nevertheless found him guilty of conspiring to defraud the Dime Savings Bank.

In July 1993, the trial judge convened a disposition hearing. Apparently fearing potential ex post facto problems, the judge, without objection, consulted the sentencing guidelines that had been in effect at the time the conspiracy wound down, namely, the June 15, 1988 edition. See United States v. Harotunian, 920 F.2d 1040, 1041-42 (1st Cir.1990) (explaining that a sentencing court should apply the guidelines in effect on the date of sentencing unless doing so will implicate ex post facto concerns); United States v. Arboleda, 929 F.2d 858, 871 (1st Cir.1991) (stating that, if the guidelines in effect at sentencing are not used, then members of a conspiracy are ordinarily "subject to the sentencing guidelines in effect at the time of the completion of the conspiracy").

Starting with a base offense level of six, see U.S.S.G. Sec. 2F1.1(a), the judge added ten levels on the theory that appellant shared responsibility for inflicting losses of at least $2,000,000 (but less than $5,000,000), see U.S.S.G. Sec. 2F1.1(b)(1)(K), and then added two incremental levels for more than minimal planning, see U.S.S.G. Sec. 2F1.1(b)(2)(A). These calculations yielded an adjusted offense level of 18, which, for a first offender, produced a guideline sentencing range (GSR) of 27-33 months. The court imposed an incarcerative sentence at the nadir of the range.

This appeal spotlights the court's determination of the aggregate losses properly attributable to LaCroix. Noting that the judge counted transactions handled by his coconspirators as "relevant conduct" under U.S.S.G. Sec. 1B1.3(a)(1), and, therefore, tagged him with the entire loss suffered by the defrauded bank, LaCroix assigns error. He contends that the sentencing court misconceived the applicable test for relevant conduct, mounted too shallow an inquiry into the subject, and, in all events, that the court found the facts in a quixotic manner, thereby misapplying the test.

Appellant's first contention poses a question of guideline interpretation, which sparks de novo review. See United States v. DeLuca, 17 F.3d 6, 7 (1st Cir.1994) (holding that, when "an appeal raises a purely legal question involving the proper interpretation of the sentencing guideline, appellate review is plenary"); United States v. St. Cyr, 977 F.2d 698, 701 (1st Cir.1992) (similar). Appellant's second contention also poses a pure question of law and is, therefore, to be reviewed under the same standard. Appellant's third contention is cut from different cloth; it hinges on a factbound determination under the applicable guideline, thus evoking clear error review. See United States v. Bradley, 917 F.2d 601, 605 (1st Cir.1990); see also United States v. Brandon, 17 F.3d 409, 458 (1st Cir.1994) (holding that valuation of losses for sentencing purposes must be reviewed under the clear error standard), petition for cert. filed (U.S. May 16, 1994) (No. 93-9135).

It is beyond serious question that the losses stemming from the 31 transactions closed by appellant constitute relevant conduct under U.S.S.G. Sec. 1B1.3(a)(1).³ Less obvious is whether the remaining transactions, approximately 60 in number, closed by coconspirators, may be attributed to him. This appeal centers around appellant's insistence that the court below misinterpreted the test governing what the Third Circuit aptly has called "accomplice attribution," see United States v. Collado, 975 F.2d 985, 990 (3d Cir.1992), by taking too permissive a view of the test's foreseeability prong.

A. The Accomplice Attribution Test.

The accomplice attribution test is restated in the case law with great frequency, but rarely in quite the same form or with quite the same emphasis. Thus, our perlustration must start with the guideline itself.

In its current iteration,4 the applicable guideline states that relevant conduct includes "all reasonably foreseeable acts and omissions of others in furtherance of the jointly undertaken criminal activity, that occurred during the commission of the offense of conviction, in preparation for that offense, or in the course of attempting to avoid detection or responsibility for that offense." U.S.S.G. Sec. 1B1.3(a)(1)(B) (Nov.1993). Reading the 1988 version of section 1B1.3(a)(1) in light of subsequent clarifying amendments to both the guideline and its commentary, we understand the Sentencing Commission to have mandated a two-part inquiry for accomplice attribution in the relevant conduct milieu. First, the sentencing court must determine what acts and omissions of others were in furtherance of the defendant's jointly undertaken criminal activity. This task requires the court to ascertain what activity fell within the scope of the specific conduct and objectives embraced by the defendant's agreement (whether explicit or tacit). Second, the court must determine to what extent others' acts and omissions that were in furtherance of jointly undertaken criminal activity likely would have been foreseeable by a reasonable person in defendant's shoes at the time of his or her agreement.5

We think it is important to emphasize that the vantage point for the foreseeability judgment is the time of the defendant's agreement--not necessarily the time he personally undertook the performance of criminal activity, or the time of his entry into the conspiracy. Siting the vantage point in this way has at least two salient implications. For one thing, a court examining relevant conduct may attribute to a defendant acts committed by his accomplices prior to the commission of his own acts, so long as they occur subsequent to his agreement. For another thing, because a single defendant may make multiple agreements or expand an existing agreement, a defendant sometimes may be chargeable with losses arising out of conduct that he could not have foreseen at the time he entered the conspiracy, so long as such conduct was foreseeable at the time that he signaled his agreement to the expanded scope of jointly undertaken criminal activity embracing such conduct.

In this case, the inquiry may be truncated. There has never been any suggestion that the 60-something transactions closed by appellant's coconspirators were outside the scope of appellant's agreement, or, put another way, that those transactions were other than in furtherance of the jointly undertaken criminal activity. Consequently, this appeal turns exclusively on the issue of foreseeability.

B. The Findings Below.

At the disposition hearing, defense counsel argued that appellant could not have foreseen the conduct of others. The lower court treated this argument as calling into question an application of the guidelines. The court then proceeded to find, based on the trial evidence and the jury verdict, that:

Mr. Lacroix was involved in this conspiracy from the beginning. He was aware of the nature and extent of the development that was involved, the development that Alpha was involved in. He was aware of the cost of the homes. He was aware of the profit that was being received, and he was also receiving salaries from Alpha, $173,000 in '87, $187,000 in '88, $25,000 in '89.

So in the opinion of the Court he was well aware of the magnitude of what was happening here, and ... under all of the circumstances in which he was involved, the foreseeability in this situation is really inherent in the nature of the conspiracy that was involved here, which was a marketing conspiracy.

Appellant says that the district court's findings on foreseeability are flawed both legally and factually.

A. Questions of Law.

Appellant raises two predominantly legal challenges to the court's formulation of the relevant conduct inquiry. We inspect each challenge in turn.

1. Mere Awareness. Appellant seizes on the district court's repeated use of the word "aware" and suggests that its recurrence betokens a single-minded focus on the defendant's knowledge. This focus is impermissible, appellant asseverates, because a finding of "mere awareness," in and of itself, cannot be equated with, and does not justify, a finding of foreseeability in the sentencing phase. Despite appellant's citations to several cases, his asseveration begs the pivotal question. Awareness does not always bear on foreseeability in precisely (or even roughly) the same way. To understand the inferences that lawfully can be drawn from awareness in any given situation, a court must first assess the particular factual setting and then answer the question: "Awareness of what?"

The four cases upon which appellant principally relies, read carefully, underscore this necessity. The first of them, United States v. O'Campo, 973 F.2d 1015, is a case in which we admonished sentencing judges not to equate mere knowledge with foreseeability--but we were referring specifically to "mere knowledge of historic facts." Id. at 1025. By this, we meant that the foreseeability of acts performed after defendant's entry into a conspiracy could not be established by his "mere knowledge" of acts performed prior to his entry into the conspiracy.6 See id. at 1026; see also United States v. Carreon, 11 F.3d 1225, 1234-37 (5th Cir.1994) (discussing O'Campo ). Since LaCroix was involved in the instant conspiracy from the start, the stratagemical acts of which he was aware necessarily occurred after his entry into the conspiracy. Thus, O'Campo is inapposite because it did not deal with post-entry acts.

The remaining three cases upon which appellant relies advance the bland proposition that foreseeability is not dispositively established by mere awareness of the existence or illegality of a conspiracy. See United States v. Evbuomwan, 992 F.2d 70, 74 (5th Cir.1993) (explaining that "mere knowledge that criminal activity is taking place is not enough"); United States v. Valencia-Lucena, 988 F.2d 228, 234 (1st Cir.1993) (suggesting that individuals may "know that the agreement they have entered is illegal but [nevertheless] have no way to foresee the magnitude or ambition of the enterprise"); United States v. Edwards, 945 F.2d 1387, 1393 (7th Cir.1991) (commenting that "foreseeability means more than subjective awareness ... that [an accomplice] headed a long-standing and successful heroin distribution network"), cert. denied, --- U.S. ----, 112 S. Ct. 1590, 118 L. Ed. 2d 308 (1992). We find no fault with these cases--but we caution that the courts' words cannot be read in a vacuum. Awareness, even if limited to knowledge of a conspiracy's unlawfulness, is always (or almost always) relevant to the question of foreseeability--and none of the cited cases suggest the contrary.

More importantly, these three cases do not in any way denigrate the possibility that foreseeability may be established by a different kind of awareness, that is, by a defendant's knowledge of the nature and extent of a conspiracy in which he is involved. This, of course, is exactly the stripe of awareness detected by the court below. It is both good law and good logic that a defendant's awareness of the inner workings of a conspiracy in which he is participating is germane to, and often highly probative of, accomplice attribution. Although appellant may choose to characterize such intimate knowledge as "mere awareness"--a term that we view as verging on the oxymoronic in a case like this one--he is fishing in an empty stream. Such knowledge frequently will suffice to prove the defendant's ability to foresee the acts of coconspirators. See, e.g., United States v. Roberts, 14 F.3d 502, 525 (10th Cir.1993) (concluding that a defendant's knowledge that the accomplice habitually carried a firearm justified a finding that defendant could reasonably foresee that accomplice would carry the gun on the occasion in question).

"Foreseeability" is conventionally defined as the "ability to see or know in advance." Black's Law Dictionary 649 (6th ed. 1990). Viewed in that light, a "reasonably foreseeable act" might well be regarded as an act that a reasonable person who knew everything that the defendant knew at the time would have been able to know in advance with a fair degree of probability. Giving due weight to the intimate connection between knowledge and foreseeability, we conclude that in this case it was both permissible and advisable for the district court to consider appellant's awareness of the conspiracy's nature and scope en route to an ultimate determination on foreseeability. The district court, therefore, did not misconstrue either the elements of the accomplice attribution test or the way in which the test should operate.

2. The Nature of the Inquiry. Appellant's next argument is pitched in a somewhat different direction. He points to a Third Circuit directive that instructs district courts, when considering accomplice attribution in the relevant conduct context, to embark upon "a searching and individualized inquiry into the circumstances surrounding each defendant's involvement in the conspiracy." Collado, 975 F.2d at 995. He then invites us to adopt this standard and calumnizes the district court for mounting an inquiry that supposedly fell short of it. We believe that this argument is largely an exercise in semantics.

In the first place, the invitation that appellant extends is wholly gratuitous. We already have endorsed the principle of a searching and individualized inquiry in the relevant conduct context. See, e.g., United States v. Balogun, 989 F.2d 20, 22 (1st Cir.1993) (holding that a sentencing court ordinarily must make specific, individualized findings regarding foreseeability for each defendant).7 Indeed, the Third Circuit, in constructing the rule appellant urges us to "adopt," cites our opinion in United States v. Garcia, 954 F.2d 12 (1st Cir.1992), as a model. See Collado, 975 F.2d at 995. The mere fact that we have employed slightly different phraseology than the Third Circuit is of no consequence. The adjectives used in Collado, while concinnous, are neither talismans nor words of art.

The second half of appellant's argument is equally meritless. Here, the district court honored the spirit of Balogun by making extensive findings regarding the foreseeability of others' acts from appellant's vantage point. Since the court presided over a 17-day trial and based its findings, inter alia, "on all of the evidence that the Court heard during the course of the trial," it strains credulity to describe the inquiry below as insufficiently searching. We are hard pressed to imagine what more the district court might have done--and appellant, for all his lamentations, has not advanced a single concrete suggestion.

B. Questions of Fact.

Appellant's fallback position is that, even if the district court applied the proper legal rules in determining relevant conduct, its findings of fact were clearly erroneous. The facts of the case, taken without embellishment, expose the fallacy in appellant's position.

To be sure, LaCroix was neither the progenitor nor the moving spirit of the conspiracy, but he helped to found it, retained a proprietary interest in it, and played an integral part in its operation. Moreover, he served as the titular head of the firm that oversaw the construction, marketing, financing, and sale of every home. The record supports--indeed, virtually compels--an inference that appellant understood from the inception that the object of the conspiracy was to sell homes by hook or by crook. Taking the district court's explicit findings, and fleshing them out with details derived from the record, we understand the court to have concluded that appellant knew all along the sums involved in each transaction and the conspiracy's method of operation--selling houses to unqualified buyers by providing, and then fraudulently concealing, secondary financing. Because any reasonable person in appellant's position, at the time of his agreement, would have recognized that ninety or more homes might be sold in this corrupt fashion, the court below did not err in concluding that all the losses resulting from the sales were fairly attributable to LaCroix.8

Appellant cannot deny this analysis in any of its particulars, and, in fairness, does not really try to do so. Instead, he seeks to escape the force of the district court's reasoning by introducing three extraneous considerations. At bottom, this endeavor reflects a basic misunderstanding of sentencing principles.

First, appellant insists that a finding of foreseeability is undermined by the jury verdict. We do not agree. The jury did not exonerate appellant in connection with the substantive offense counts; rather, it simply deadlocked on these counts. Its verdict, therefore, did not resolve the contested issues either way, but left them up in the air.

Moreover, the method of the guidelines is to leave to the sentencing judge, not the jury, the determination of what "conduct" is "relevant" to the fashioning of a defendant's sentence. See United States v. Limberopoulos, 26 F.3d 245, 251-52 (1st Cir.1994); see also U.S.S.G. Sec. 6A1.3. Thus, even a trial jury's refusal to find that a certain fact has been proven beyond a reasonable doubt will not bar the district court from making precisely that same finding at sentencing, under a preponderance-of-the-evidence standard. On this basis, we have held squarely that a defendant's acquittal on a particular count does not limit the sentencing court's flexibility in considering the same underlying facts in respect to the count of conviction. See United States v. Mocciola, 891 F.2d 13, 16 (1st Cir.1989). A fortiori, the case for permitting judges free rein to make whatever findings the record can support is airtight where, as here, trial on the disputed counts ends with a hung jury rather than an acquittal.

Second, appellant presents himself as a babe in the woods, an uneducated carpenter in the company of sophisticated entrepreneurs. For what this jeremiad may be worth insofar as it bears upon accomplice attribution, it was tendered to, and rejected by, the district judge.9 We discern no clear error. See, e.g., United States v. Ruiz, 905 F.2d 499, 508 (1st Cir.1990) (acknowledging that "where there is more than one plausible view of the circumstances, the sentencing court's choice among supportable alternatives cannot be clearly erroneous").

Third, and relatedly, appellant harps on the fact that Zsofka and Horn called the tune, to which he merely danced. But the concepts of "relevant conduct" and "role" are distinct in the world of the sentencing guidelines. See United States v. Lilly, 13 F.3d 15, 18-19 (1st Cir.1994). Whereas the former helps to gauge the gravity of an offense, the latter helps to measure the offender's culpability. See id. Hence, the district court's attribution of the entire loss to appellant is not in any way inconsistent with the fact that he may have played only a supporting role.10

We need go no further. The short of it is that none of the factors upon which appellant dwells cast doubt upon the district court's ascertainment of the amount of loss attributable to appellant in connection with the jointly undertaken criminal activity. Consequently, we cannot say that the lower court erred in constructing the sentencing calculus.

Affirmed.