Sears, Roebuck & Company, Petitioner/cross-Respondent v. National Labor Relations Board, Respondent/cross-Petitioner , 349 F.3d 493 ( 2003 )

MANION, Circuit Judge.

Sears, Roebuck, and Company petitions for review of a decision of the National Labor Relations Board (“NLRB” or “Board”) holding that Sears violated 29 U.S.C. §§ 158(a)(1) and (a)(3) by firing John Iaci, Corliss Hepburn, and Cordy Richardson for their protected union activities. Sears requests that we deny enforcement of the Board’s order that the three be reinstated and otherwise compensated. The General Counsel of the NLRB cross-petitions for enforcement. Because substantial evidence does not support the Board’s decision, we grant the petition for review and deny enforcement.

I.

At the end of 1996, John Iaci was an appliance repairman for Sears in West Palm Beach, Florida. Iaci had three decades of employment with Sears and, although Iaci had occasionally received what the General Counsel characterizes as “minor, informal discipline,” his performance evaluations were mostly positive. The performance reviews, and possible annual raises, of a repairman like Iaci depended partly on the number of service calls that he completed. In February 1997, Iaci claimed falsely to have completed two repairs at a customer’s house when, in truth, he had not even visited the customer’s residence. Iaci’s immediate supervisor, Christine Smith, discovered that he had falsified company records to reflect that he had completed the service calls and reported the problem to her boss, District Service Manager Ron Medford. In her memorandum to Medford, Smith stated that “Mr. Iaci is manipulating the system to his own benefit and to the customers [sic] detriment. In the time Mr. Iaci is doing these ‘dummy’ rounds he could actually be handling customers who need service.” She also noted that Iaci’s misconduct had delayed customer repairs, added burdens to the repairmen who actually were visiting all of the customers whom they claimed to visit, and forced Sears to pay overtime.

Medford gave Iaci a chance to respond to Smith’s conclusions. In a three-page, hand-written memorandum, Iaci threw himself on Medford’s mercy. Iaci wrote that “I am very sorry 4 what I did + know it was wrong.” He apologized profusely for his “mistake,” admitting that he “said he had checked” a customer’s dishwasher and washer “when in fact [he] never went” to the customer’s residence. Iaci also pointed out, however, that he had “given [his] life to Sears” and that he had needed to work because he had a wife and house. He implored Medford, “[p]lease do not terminate me” and ended his memorandum with the following:

You probably think well if he did it once how many times before he has done it. I understand that. Please call [the customers on] my routes you will see this is not happening. This was isolated. I don’t know what else to say except that I’m sorry. Very sorry.

Under the “Sears Human Resources Guide for Managers,” “falsification of Company records” and “improper recording of detail” were grounds for “immediate termination.” Instead of firing Iaci immediately, however, Medford decided to suspend him for a few days. After Iaci returned to work, Smith followed Iaci’s suggestion and paid particular attention to his routes. Medford, for his part, warned Iaci that further violations of company policy would lead to termination of his employment. The Board aptly characterized this as a “last chance warning.”

Shortly after his return to work, Iaci became involved with the International Brotherhood of Electrical Workers, Local Union 349, AFL-CIO (“the Union”). According to Iaci’s testimony, he had learned in late 1996 or early 1997 that the Union was attempting to organize Sears’s West Palm Beach facility. Iaci started to play an active role in this attempt in the spring of 1997, when he began distributing union authorization cards to, and discussing the Union with, coworkers in the facility’s parking lot. As Iaci put it, he distributed “approximately fourteen cards to fourteen different employees” in “[l]ate April, maybe through May” 1997.¹ In June 1997, Medford had a conversation with Iaci. Medford stated that Sears did not want a “third party” at the West Palm Beach facility, and he then asked Iaci if transferring Smith to a different location would “make the third party go away.” Iaci replied that he did not know whether Smith’s departure would make the “third party” go away, but that it would not hurt. Later that month, Medford assumed a new position with Sears and relocated to Illinois.

In the meantime, according to Smith’s testimony, Iaci’s troubles with Smith continued. Smith testified that, in March and April, she documented six instances in which Iaci falsely recorded that he had repaired a customer’s appliance when, in fact, the customer had declined the repair. Again on July 28, August 4, and August 15, according to Smith’s testimony, Iaci falsely reported that he had repaired an item when, in fact, he had not. (Ironically, Smith testified that she verified the infractions of July 28 and August 15 by, as Iaci himself had suggested, calling the customers on his routes.) Smith testified that, by lying about these events, Iaci once again falsely inflated his record of productivity; where no warranty or maintenance contract applied, Sears unsurprisingly charged customers more for a repair than it charged merely for sending a repairman to the customer’s house and providing an estimate. Although Smith testified that she had counseled Iaci orally about these reports, from March through July, nonetheless, her monthly evaluations of Iaci’s work were positive, containing compliments like “great job” and “this is a great performance.”

In mid-August, Ralph Graettinger replaced Medford as the District Service Manager for the West Palm Beach and Plantation facilities. On August 21, Graet-tinger had a meeting with Iaci about issues related to work. Iaci told Graettinger that he was trying to straighten out some of the problems at work and that he was upset about the Union. According to Iaci’s testimony, Iaci had said that “most of the employees really didn’t want a Union but they did want the problems straightened out.” Graettinger then stated to Iaci that he was not going to transfer Smith, and that Iaci had a “bad attitude,” was too opinionated, was a bad influence on other employees, and said things that other employees should not hear.

In late August, according to Smith’s testimony, Smith showed Graettinger the documentation regarding Iaci’s false reports of having repaired appliances. Gr.aettinger replied that Smith should prepare a chronological summary of Iaci’s false reports. During her investigation, according to Smith’s testimony, she discovered three more discrepancies on Iaci’s routes, one of which involved an occasion on which Iaci had changed the warranty date on a customer’s washing machine so that Sears (and not the customer) would assume the cost of the repair to that machine. Smith testified that Iaci then sold that same customer a maintenance agreement, on which Iaci reaped a commission. Smith testified that she presented a memorandum with supporting documentation to Graettinger on September 15, concluding that Iaci “continued to falsify his routes knowing the penalty for repeating these actions would be cause for dismissal.”

Graettinger then had Sears’s loss prevention personnel, under the direction of Richard Gonzalez, interview Iaci and investigate whether there was a legitimate explanation for his conduct. During the interview, Gonzalez assured Iaci that he would not be fired and essentially dictated the last two sentences of the statement that Iaci wrote.² That statement reads as follows:

On this date I discussed with Mr. Gonzalez about picking up calls while at the house and doing a minimal check. I was doing this to generate more completed calls and also at times to satisfy a customer request. I did not know Sears did not want this done. I know now and will not do this unless cust needs repair. Sometimes I did not do any check. On some occasions I changed the warranty date and sold an MA. The customer would get call done free and MA was sold. I understand doing this cause a loss to the company.

Immediately after reading the statement, Graettinger entered the interview room and fired Iaci.

Like John Iaci, Cordy Richardson was an experienced repairman for Sears, having worked for the company for more than two decades. In 1997 and 1998, Richardson worked out of Sears’s facility in Plantation, Florida, and was also under the indirect supervision of Graettinger. Cor-liss Hepburn worked out of the same location, having repaired sewing machines and vacuum cleaners for about 14 years. She too was under Graettinger’s indirect supervision. As was the case with Iaci, both Richardson and Hepburn were involved in the Union’s attempt to organize the workforce. Each spoke favorably about the Union during a meeting held at the Plantation facility sometime during the week of February 23, 1998³ that was called by Sears and that was attended by at least one Sears supervisor, Charlie Young, who argued against unionization. Both Richardson and Hepburn solicited union authorization cards. In addition to his union activities, Richardson also regularly ate lunch at a local restaurant with Joe Fowler, a former Sears employee who had retired in May 1997 and who was instrumental in the attempt to organize Sears’s facilities at Plantation and West Palm Beach. On one occasion in March 1997, Sears supervisor Pat McLaughlin and Sears employee James Eassey watched the restaurant with binoculars to see who was there.

In late February 1998, shortly before the representation election on February 27,⁴ Fowler distributed an unsigned letter dated February 20. The letter espoused the virtues of the Union to Sears’s employees at the Plantation facility, was addressed to “Dear Fellow Worker,” and began with the phrase ‘Why I will vote in favor of the Union.” As Fowler testified, he “wanted it [the letter] to sound as if it was coming from someone who still worked” for Sears. (Fowler, of course, was no longer a “Fellow Worker” at Sears, and he was therefore in no position to “vote in favor of the Union.”) Unsurprisingly, speculation arose as to the authorship of Fowler’s anonymous letter. Sandra Smith, who stated that she was “in charge of cashier audit and data entry” in February 1998 at Sears’s Plantation facility, testified that, although she did not know who wrote the letter, “there was a whole lot of gossip” circulating at the Plantation facility regarding the letter.

During one of her days off in mid-February 1998, Hepburn, who owned a Sears refrigerator, called the service center to request a repair, specifically asking that Richardson be assigned to fix the appliance. There was nothing necessarily unusual about that request: employees who were also owners of Sears appliances were treated the same as any other customer needing a repair, and sometimes a customer requested a specific repairman by name. Richardson went to Hepburn’s residence and replaced the compressor in Hepburn’s refrigerator. He also, however, incorrectly reported that Hepburn’s refrigerator was covered by a “service flash” that Sears had issued as to 24 model numbers. The service flash extended the warranty period for models that had a tendency to break within an abnormally short amount of time. Although Hepburn’s refrigerator was of a similar age and type to some of the 24 models listed in the service flash, it was not actually covered. The effect of Richardson’s incorrect report, had Sears not discovered it, would have been that Hepburn would have improperly received a free repair and compressor, which together were worth about $600.

Thanks to an alert auditor, however, Sears learned within days of the repair that Richardson had incorrectly reported that Hepburn’s refrigerator was covered by the service flash. Graettinger testified that, upon hearing of the incident, he immediately concluded that the two had conspired to steal a compressor from Sears, and he thus sought permission to fire them. Graettinger received that permission, subject to one condition: he had to prove that his suspicions were correct. Toward that end, he had Gonzalez interview separately Richardson and Hepburn on March 6, 1998, which was about one week after employees voted 139 to 62 against the Union. During his interview with Richardson, Gonzalez told him that his “mistake was not serious” and that the purpose of the interview was a “slap on the wrist.” Richardson then wrote the following statement, which was virtually dictated ⁵ to him by Gonzalez:

On 2/17/98 I Replaced A Compressor In A Associate Refrigerator And Put It In As Customer Satisfaction. I Was To Charge For Part And Labor Out Of Warranty 1 Year. Total Cost Would Have Been Around 600.00. This Is What I Should Have Charge Her. I Didn’t Because I Trying To Be Nice. I Understand The Company Loses Money When I Do This I Know It Is Wrong. I Didn’t Receive Any Money On The Side For This.

Similarly, Gonzalez interviewed Hepburn. She wrote the following statement after Gonzalez essentially told her what to write:

2 weeks ago Tech Richardson came to my home, repaired my refrigerator which was out of warranty and he didn’t charge me for the repair. Doing me a favor. The charge was $600.00. He didn’t charge me anything. No money was paid to tech Richardson, I offered to buy him lunch which he refused. I now realize this is wrong. At the time I didn’t think of it like that, but now I know that its wrong and that the Company loses money.

Statements in hand, Graettinger first met with Richardson. Graettinger told Richardson that his conduct was tantamount to stealing from Sears and that he was fired. Graettinger then met with Hepburn and fired her as well. Hepburn testified that, when she asked Graettinger whether she was being fired because of the Union, he answered no and smiled. Graet-tinger testified that he had not smiled and that Hepburn never asked him whether her discharge was for protected activities.

The Board’s General Counsel later filed an administrative action, alleging that Sears violated 29 U.S.C. §§ 158(a)(1) and (a)(3) by firing Iaci, Richardson, and Hepburn for their union activities. The ALJ, Jane Vandeventer, agreed, ordering Sears to reinstate, and otherwise make whole, the three workers. The Board upheld the ALJ’s order on administrative appeal, although it limited its discussion to Iaci’s firing. See Sears, Roebuck & Co., 337 NLRB No. 65, 2002 WL 736260 (2002).

n.

We have jurisdiction to review, pursuant to 29 U.S.C. §§ 160(e) and (f), petitions for review of Board decisions. In conducting our review, we must determine whether “the Board’s decision is supported by substantial evidence and whether its legal conclusions have a reasonable basis in law.” International Union of Operating Engineers v. NLRB, 325 F.3d 818, 828 (7th Cir.2003). The substantial evidence standard applies to the Board’s factual findings, and is satisfied when the Board relies upon “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Id. “[W]e defer particularly to the Board’s findings regarding credibility, which cannot be disturbed absent extraordinary circumstances.” Id. As to whether the Board’s legal conclusions have a reasonable basis in law, we must apply a similarly deferential standard. Id. With these standards in mind, we turn to the relevant statutory provisions.

Under 29 U.S.C. § 157:

Employees shall have the right to self-organization, to form, join, or assist labor organizations, to bargain collectively through representatives of their own choosing, and to engage in other concerted activities for the purpose of collective bargaining or other mutual aid or protection, and shall also have the right to refrain from any or all of such activities except to the extent that such right may be affected by an agreement requiring membership in a labor organization as a condition of employment as authorized in section 158(a)(3) of this title.

These rights are protected by 29 U.S.C. § 158. Section 158(a)(1) makes it unlawful for a covered employer “to interfere with, restrain, or coerce employees in the exercise of rights guaranteed in section 157 of this title,” and § 158(a)(3) makes it unlawful for a covered employer to discriminate against employees regarding tenure of employment in order to discourage membership in a labor union. Van Vlerah Mechanical, Inc. v. NLRB, 130 F.3d 1258, 1262-63 (7th Cir.1997). The Board ordered Iaci, Richardson, and Hepburn reinstated because it concluded that Sears violated both provisions by firing them because of their union activities.

If a covered employer were to fire an employee because of his union activities, it would trigger liability under both §§ 158(a)(1) and (a)(3). Jet Star, Inc. v. NLRB, 209 F.3d 671, 675 (7th Cir.2000). To establish a prima facie case that an employer has committed this offense, the Board’s General Counsel must show that: (1) the employee engaged in protected activity; (2) the decisionmaker knew it; and (3) the employer acted because of anti-union animus. Id. If the General Counsel establishes a prima facie case, the employer then must either rebut the General Counsel’s evidence or prove, pursuant to the affirmative defense delineated in Wright Line, a Din. of Wright Line, Inc., 251 NLRB 1083, 1980 WL 12312 (1980), that the employment action would have been taken in the absence of the protected activities. Id. The prima facie case and the affirmative defense available under Wright Line are linked: the weaker the prima facie case, the easier it is for the employer to establish that it would have taken the adverse action regardless of the employee’s protected activity. Sasol N. Am., Inc. v. NLRB, 275 F.3d 1106, 1113 (D.C.Cir.2002).

A. John Iaci

We turn first to John Iaci. Iaci was involved in union activities in April or May 1997 when he distributed authorization cards and promoted the Union to his fellow employees. NLRB v. Q-1 Motor Exp., Inc., 25 F.3d 473, 478 (7th Cir.1994). It is less clear that substantial evidence supports the second element of the prima facie case, knowledge of Iaci’s protected activities. To satisfy this element, the General Counsel had to show that the decisionmaker(s) responsible for the firing of Iaci knew that he was involved in union activities. Vulcan Basement Waterproofing v. NLRB, 219 F.3d 677, 685 (7th Cir.2000) (reasoning that “the decision-makers at Vulcan had to know of’ the protected activities); Jim Walter Resources, Inc. v. NLRB, 177 F.3d 961, 963 (11th Cir.1999) (reasoning that “[n]one of the persons who gave ‘bad attitude’ statements was ‘an agent responsible for hiring’ ”); NLRB v. McEver Eng’g, Inc., 784 F.2d 634, 640 (6th Cir.1986) (holding that “[b]efore an employer can be said to' have discriminated against its employees for their protected activity, the Board must show that the supervisor responsible for the alleged discriminatory action knew about the” union activity); Air Surrey Corp. v. NLRB, 601 F.2d 266, 257-58 (6th Cir.1979) (vacating the Board’s order because substantial evidence did not show that the employee’s supervisor knew of his protected activity); see also Richdel, Inc., 265 NLRB 467, 475-75, 1982 WL 24028 (1982) (focusing on the decisionmaker’s knowledge of protected activities). In holding that the General Counsel had met his burden, the Board reasoned thus:

[W]e find, contrary to [Sears’s] argument, that the record contains ample evidence of employer knowledge of Iaci’s union activity. The judge credited Iaci’s testimony concerning conversations with District Service Managers Medford and Graettinger in June and August, respectively. In the June conversation, Iaci was asked whether the “third party” “would go away” if Supervisor Smith were transferred. The term “third party” clearly referred to the Union. Thus, it is evident that [Sears] was aware that at the time that Iaci was involved with the Union. Graettinger’s reference in the August conversation to [Sears’s] decision not to transfer Smith indicates that Graettinger was aware of the content of Iaci’s June conversation with Medford. In addition, Iaci specifically mentioned the Union in the August conversation. The judge discredited Graet-tinger’s denial that he knew about Iaci’s union activities. We find the record sufficient to support a finding that [Sears] had knowledge of Iaci’s union support and activities.

We agree that the record demonstrates that Medford and Iaci had a conversation about the Union. Medford, however was not the decisionmaker who fired Iaci; Graettinger made that decision.⁶ Therefore, Medford’s knowledge of Iaci’s involvement with the Union is relevant only insofar as it may be imputed to Qraettinger. Essentially, the Board imputed that knowledge to Graettinger because it reasoned that Medford had discussed with Iaci the possibility of transferring Smith in order to foil the drive toward unionization, Medford told Graettinger that Iaci might be interested in Smith’s transfer and, therefore, it was probable that Graettinger was aware of the antiunion context within which Smith’s possible transfer was broached. Considering that Iaci testified that Graettinger had told him that “Chris Smith’s name had come up with Mr. Med-ford, and he told me that she was not going to be moved,” a reasonable person could conclude that Medford and Graet-tinger had spoken, and that Iaci’s involvement with the Union was a topic of conversation. Although this statement was a thin reed to support the element of knowledge, we nevertheless hold that substantial evidence supports the Board’s finding that Graettinger knew of Iaci’s protected activities.

We now turn to prong three of the prima facie case and must decide whether substantial evidence supports the Board’s finding that Graettinger acted because of antiunion animus. The Board identified three reasons upon which it based its finding of animus. The Board stated that, “[in] finding that the record supports an inference of animus, we rely on the timing of Iaci’s discharge and the ‘blatant disparity’ between the treatment of Iaci and that of other employees who engaged in similar work infractions.” In a footnote, the Board added its third reason, stating that “[i]n addition, we agree with the judge that Medford’s questioning of Iaci as to what steps [Sears] might take in order to make the Union ‘go away’ shows antiunion animus.”

We begin with the timing of Iaci’s discharge. Coincidence between union activity and an employee’s discharge may, when added to other evidence of an employer’s antiunion motivation, form part of the substantial evidence underlying a finding of antiunion animus. Chicago Tribune Co. v. NLRB, 962 F.2d 712, 717-18 (7th Cir.1992).⁷ The coincidence relied upon by the Board, however, is exceedingly weak. Iaci’s union activities consisted of distributing about 14 union authorization cards to, and discussing the Union with, co-workers in April or May 1997. Sears did not fire Iaci, however, until October 2, 1997, more than four months after his protected activities had ended. There is thus a significant lapse in time between Iaci’s protected activities and his discharge. Cf. NLRB v. Stor-Rite Metal Products, Inc., 856 F.2d 957, 965 (7th Cir.1988) (asking rhetorically, “[i]f Stor-Rite acted with retaliatory intent, then why did it delay the full impact of its retaliation until months” after the protected conduct?).

Further undermining the Board’s reb-anee on the timing of Iaci’s discharge is that the string of misconduct for which Sears first disciplined, and later purported to fire, Iaci began well before the union election and well before Iaci’s protected activities. See Chicago Tribune Co., 962 F.2d at 718 (noting that “Kaczmarek’s poor work record dated to well before the union election”); Peavey Co. v. NLRB, 648 F.2d 460, 462 (7th Cir.1981) (noting that “it is undisputed that Snider had been disciplined, for cause, prior to her contact with the union”). It is undisputed that Iaci was disciplined and nearly fired in February 1997 for filing reports, in which he falsely claimed to have fixed appliances, months before he distributed authorization cards. The testimony of Christine Smith, testimony that the ALJ and the Board never acknowledged, was that Iaci continued to file reports, on various occasions from March through mid-August 1997, in which he falsely claimed to have repaired appliances. Smith testified that she had orally counseled Iaci about these infractions repeatedly. Iaci himself admitted in his written statement of October 2, 1997 that “[s]ometimes I did not do any check.” (Although the ALJ found that the last two sentences of Iaci’s statement were dictated by Gonzalez, she made no such finding as to the preceding statement, in which Iaci confessed to the same conduct for which he had been disciplined in February 1997.)

Where, as here, an employee’s discharge purportedly stems from a series of disciplinary incidents or warnings that predate the employee’s union activities, the timing of that discharge rarely if ever constitutes substantial evidence of the employer’s antiunion animus. See id.; see also NLRB v. Newman-Green, Inc., 401 F.2d 1, 3-4 (7th Cir.1968) (holding that no substantial evidence supported the Board’s finding of anti-union animus where the employee, who had been repeatedly disciplined for coming to work under the influence of alcohol, was fired for drunkenness); 1 The Developing Labor Law 297 (Hardin et al. eds., 4th ed.2001) (reasoning that “the giving of warnings for specific conduct may suggest that a subsequent discharge based upon similar conduct is not discriminatorily motivated”). This case is no exception to the rule. Iaci had begun the series of infractions for which Sears ostensibly fired him well before he began his protected activities. And he had been counseled, if not warned, continually as these infractions occurred. In light of that consideration, no reasonable person could conclude that, simply because Sears fired Iaci more than four months after his union activities had ceased, it had therefore fired him because of his union activities. The temporal link between lad’s protected activities and his firing is, as we said in Chicago Tribune Co. v. NLRB, “too remote, indeterminate and ethereal” to amount to substantial evidence of antiunion animus. 962 F.2d at 718; cf. Medeco Sec. Locks, Inc. v. NLRB, 142 F.3d 733, 743-44 (4th Cir.1998) (holding that no substantial evidence of knowledge could be inferred from the company’s decision to fire a worker within a year after he had ceased to be a visible supporter of the union and had disavowed further interest in the union). To reach the contrary conclusion, and hold that substantial evidence of antiunion animus exists merely when an employer knows of a worker’s union activities and later fires that employee, would be tantamount to making union activism a shield against discharge, which is a result that would be incompatible with the statute. Loy Food Stores, Inc., 697 F.2d at 801.

Next, we look to the second reason that the Board found antiunion animus: “the ‘blatant disparity’ between the treatment of Iaci and that of other employees who engaged in similar work infractions.” The Board is correct in observing that the disparate disciplinary treatment of employees who engaged in union activities can constitute substantial evidence of anti-union animus. Great Lakes Warehouse Corp. v. NLRB, 239 F.3d 886, 891 (7th Cir.2001). Despite opining that the discrepancy in treatment in this case was “blatant,” however, the Board did not identify a single employee who, like Iaci, falsified reports of service calls, but was treated differently than Iaci was treated. We would expect that, if the disparate treatment were really “blatant,” or even if there were just substantial evidence of disparate treatment, the Board would be able to point to at least one specific employee who was disciplined more lightly for infractions similar to Iaci’s misconduct. In Great Lakes Warehouse Corp. v. NLRB, for example, the Board’s decision as to disparate treatment rested on evidence that the employer “had been more lenient in the application of its disciplinary policy” toward specifically named employees who were less identified with the union. Id. at 889-91.

Here, however, the Board identifies no such employee. Moreover, what is most perplexing about the Board’s silence on this point is that Sears does point to one other employee who committed basically the same conduct for which Iaci was fired. According to the uncontradicted testimony of Christine Smith, Bruce Edwards, whom Sears had employed since 1974, was another senior repairman under her supervision who, in March 1997, falsely reported that he was conducting a service call during a time in which he was actually at his own home. As she had done in Iaci’s case, Smith reported the problem to Medford, who asked Edwards to respond. Edwards admitted his misconduct and wrote a confession. Medford then suspended and demoted Edwards, and warned him that further misconduct of a similar nature would result in immediate discharge. When Edwards returned to work, Smith scrutinized his routes more carefully. In mid-September 1997, when Smith learned that Edwards had changed the warranty date on an appliance that he owned, and then installed Sears’s parts for free, she reported the problem to Graettinger. After Gonzalez interviewed Edwards on October 2, 1997, and Edwards admitted his misconduct, Graettinger fired him.

Edwards’s situation was almost a mirror image of Iaci’s situation. In early 1997 Medford suspended both, and gave both a last chance warning, for falsifying reports of their service calls. After confessing his misdeeds, each was allowed to return to work under Smith’s increased scrutiny. Both then falsified warranty dates, Iaci on a customer’s appliance, Edwards on his own. Had Sears not detected both falsifications, the effect of each would have been that Sears would have unwittingly given away parts and repairs for which a customer should have paid. After Graettinger became aware of Iaci’s and Edwards’s misconduct, he fired both in October 1997 after brief investigations by Gonzalez. There is only one glaring difference between Iaci and Edwards: in the case of Edwards, there is no evidence that he had any involvement with the Union.

To refute the General Counsel’s allegation of disparate treatment, Sears cited the example of Edwards before the Board. There are strong parallels between Sears’s investigation and firing of Edwards and its investigation and firing of Iaci. Yet the Board (and the ALJ, for that matter) did not even mention Edwards’s name before concluding that there was a “blatant disparity” between Sears’s treatment of Iaci and other employees who had committed similar misconduct, but were not involved with the Union. It would have been error merely for the Board to select and discuss only the evidence that favored its conclusion (that Sears fired Iaci because of the prohibited animus), while failing to articulate its reasons for rejecting a line of countervailing evidence. Herron v. Shalala, 19 F.3d 329, 333 (7th Cir.1994). What the Board did in this case, however, was worse. The Board not only ignored Sears’s Edwards-related evidence; it also failed to select and discuss the evidence that would have shown that Sears treated at least one similarly-situated employee differently than it treated Iaci. We therefore hold that substantial evidence did not underlie the Board’s disparate-treatment theory.

We next address the final reason that the Board found animus regarding Iaci: Medford’s questioning of Iaci about how to make the “third party” “go away.” We agree with the Board that this conversation could be construed as substantial evidence of Medford’s antiunion animus. But Medford was not involved in the decision to fire Iaci and his antiunion animus, therefore, is irrelevant. What matters is the antiunion animus of Graettinger, the decision-making supervisor who fired Iaci. See Vulcan, 219 F.3d at 686.⁸ Or, as the attorney for the General Counsel aptly put it during oral argument, “the question needs to be whether or not he [Graettinger] had bias.” Medford’s conversation with Iaci sheds no light on the issue of whether Graettinger acted out of ill will toward the Union, and therefore does not constitute substantial, or even relevant, evidence that antiunion animus caused Iaci’s discharge.

The Board provided three reasons for its conclusion that Sears terminated Iaci’s employment because of antiunion animus: (1) the timing of the discharge; (2) the “blatant disparity” between Sears’s treatment of Iaci and its treatment of employees who committed similar offenses, but were not involved in protected activity; and (3) Medford’s conversation with Iaci about union activities. Because the evidence underlying these three reasons, taken individually or as a whole, does not constitute substantial evidence of Graet-tinger’s animus, we refuse to enforce the Board’s order insofar as it pertains to Iaci.

We also refuse to enforce the Board’s order as to Iaci on an alternate ground: Sears had shown that it would have fired Iaci regardless of his protected activity, and the ALJ relied upon no substantial evidence when she reached the contrary conclusion. In its brief before the Board, Sears contended that it was entitled to the affirmative defense provided by Wright Line because it had fired Iaci for falsifying his production records by recording service calls he did not actually make and also because Iaci had changed the warranty date on a product to sell the customer a maintenance agreement. The Board rejected this argument without conducting its own analysis, stating that “[f]or the reasons set forth by the [ALJ], we agree” that Sears had not proven that it would have fired Iaci even without his union activities. On this point, we therefore review the ALJ’s decision for substantial evidence. NLRB v. Federal Sec., Inc., 154 F.3d 751, 755 (7th Cir.1998).

The ALJ’s first reason for rejecting Sears’s position was, essentially, that Sears had not fired a similar employee for committing the same offenses that Iaci had perpetrated, and that its claim of having discharged Iaci for filing false reports was therefore pretextual. The ALJ explained that Sears

had introduced no evidence of any other employee of comparable seniority, 31 years, who was fired for doing service checks on a second appliance, or for changing warranty dates, or selling a maintenance agreement. The evidence of allegedly consistent discharges upon which [Sears] relied, those employees interviewed by investigator Gonzalez in the same time period, were discharged for garden variety theft, either of appliance parts or of money. Respondent introduced no evidence of any kind or of even minor discipline, much less discharge, for the same type of alleged infractions on the basis of which [Iaci] was allegedly discharged. Except for cases of ordinary theft, and in the cases of Hepburn and Richardson, [Sears] can point to no other employees who were interviewed by Gonzalez only, without any investigation by a service supervisor.

The ALJ attacked two straw men. Sears never argued that Iaci’s dismissal was for “doing service checks on a second appliance ... or selling a maintenance agreement.” It actually contended that it had fired Iaci for falsifying his production records by recording service calls he did not really make and also because Iaci had changed the warranty date on a product to sell the customer a maintenance agreement. Furthermore, the ALJ’s conclusion that Sears “had introduced no evidence of any other employee of comparable seniority, 31 years, who was fired for doing” the same thing that Iaci did was simply incorrect. As discussed above, Sears submitted evidence that it had fired Edwards, whose seniority was comparable to Iaci’s (23 years versus 31 years), for the same misconduct for which it discharged Iaci: falsifying reports of service calls and a warranty date. As was true of Iaci, Edwards was interviewed by Gonzalez before he was fired. The ALJ, however, never mentioned Edwards’s name, much less explained why she was rejecting that line of evidence.

We now turn to the second and final reason that the ALJ gave for rejecting Sears’s affirmative defense: that Iaci did not repeat the actions for which Sears had given him a “last chance warning” in February 1997. The ALJ reasoned thus:

The evidence likewise shows that [Iaci] did not repeat his transgression of February 1997, which consisted of recording as “complete” [a] service call that had been cancelled before he had visited the location. The alleged transgressions upon which [Sears] relied in September and October 1997 were different, and were shown by testimony of neutral and reliable witnesses to be common practices, rather than egregious sins against [Sears’s] policy. [Sears] could point to no rules specifically prohibiting these practices. Even assuming that these practices technically violated a policy of [Sears], they were not shown to be cause for discipline of any kind among other service technicians.

The essence of the ALJ’s reasoning was that, because Iaci’s transgressions of September and October were different from the conduct to which Iaci had admitted in February 1997 (falsely reporting that he had repaired items that he had not repaired), and because Iaci’s misconduct of September and October was not serious, the evidence proved that Iaci “did not repeat his transgression of February 1997.” In reaching this conclusion, however, the ALJ ignored Smith’s extensive testimony that, from March through August 1997, Iaci repeatedly filed false reports, claiming to have fixed appliances that he had not' actually repaired.⁹ In other words, the ALJ ignored evidence that Iaci actually had repeated the very conduct that had caused Sears to issue its “last chance warning” in February 1997. This was error, because the ALJ was obligated to consider all relevant evidence, not just the evidence that favored her ultimate conclusion. Clifford v. Apfel, 227 F.3d 863, 871 (7th Cir.2000). We therefore hold that the ALJ’s second reason for rejecting Sears’s affirmative defense, that Iaci “did not repeat his transgression of February 1997,” was not supported by substantial evidence. On the contrary, Sears had submitted ample evidence that Iaci had repeated his transgressions of February 1997 on several occasions during the summer. Accordingly, we refuse to enforce the Board’s order as to John Iaci on the alternate ground that Sears is entitled to the affirmative defense provided by Wright Line.

B. Corliss Hepburn

As we discussed above, for us to enforce the Board’s order as to each individual employee, substantial evidence must support the Board’s conclusion that the deci-sionmaker who discharged the employee knew that the fired employee was involved in protected activities. As to Corliss Hepburn, the Board adopted the ALJ’s opinion regarding this element of the prima facie case. We therefore look to the ALJ’s decision in search of substantial evidence that the decisionmaker, who in this case was Ralph Graettinger, knew that Hepburn was involved with the Union before he decided to discharge her. The ALJ provided three reasons for concluding that the element of knowledge existed: (1) Hepburn’s attendance and remarks at the meeting held during the week of February 23, 1998; (2) the “widespread belief’ that Hepburn was the author of the unsigned letter that Joe Fowler had written and circulated in late February 1998, just before the union election; and (3) “Graettinger’s reaction to Hepburn’s mention of the Union after her discharge.” We begin by emphasizing again that the dispositive inquiry is whether substantial evidence supports the view that the decisionmaker knew of the employee’s union activities.

Regarding the ALJ’s first reason for finding the element of knowledge, Hepburn’s remarks at the meeting held during the week of February 23, 1998, the ALJ wrote that “[t]he fact that Hepburn made remarks at the mid-February 1998 meeting called by Respondent [i.e., Sears] and addressed by Charlie Young which would be construed as critical of Respondent is sufficient to show that Respondent had notice of her pro-union sentiments .... In addition, the fact that she and Richardson were both assigned to attend the Respondent-called meeting at which all the employees were Union supporters is also persuasive [sic] that Respondent knew that they were both Union supporters.” Hepburn’s remarks are relevant to the issue of knowledge only to the extent that they show that the decision-maker, Graettinger, was aware of Hepburn’s protected activities. Vulcan, 219 F.3d at 685. For the ALJ to focus simply on the question of whether “Respondent,” i.e., Sears, knew of Hepburn’s comments was therefore not a reasonable application of the statute. For purposes of the prima facie case under §§ 158(a)(1) and (a)(3), an employer can only be said to know of the employee’s protected activities through the decisionmaker. See id. As to Graettinger, the decisionmaker who fired Hepburn, nothing in the ALJ’s opinion (or the record, as far as we can tell) provides any indication that he knew that Hepburn made any pro-union comments at the meeting. Hepburn’s testimony was that Graettinger made an appearance at the meeting only after she had finished speaking. Hepburn’s comments at the meeting, therefore, cannot constitute substantial evidence that Graettinger knew that she had engaged in protected acts.

We also observe that no evidence cited in the ALJ’s opinion suggests that Graettinger was aware of the fact that employees attending the meeting were, ipso facto, in favor of the Union. Moreover, even if Graettinger had known that employees attending the meeting were sympathetic to the Union, that knowledge alone would not constitute substantial evidence that Graettinger knew that Hepburn was engaged in a protected activity. Sections 158(a)(1) and (a)(3) protect, as we discussed earlier, employees’ rights to engage in union activities. Van Vlerah Mechanical, Inc., 130 F.3d at 1262. There is no evidence that any employee’s attendance at the meeting was such an activity. Rather, as the General Counsel himself characterizes it, the meeting was a “mandatory anti-union meeting.” An employee’s compelled attendance at an anti union meeting is not an activity that §§ 158(a)(1) and (a)(3) protect. Because Hepburn’s compelled attendance at the meeting was not protected activity, Graettinger’s ostensible knowledge of her attendance and of the pro-union sympathies of the employees present is irrelevant.

We turn next to the second reason the ALJ gave for finding the element of knowledge regarding Hepburn: the “widespread belief’ that Hepburn was the author of the unsigned letter that Joe Fowler had written and circulated in late February 1998, just before the representation election. The ALJ points to no direct evidence that Graettinger himself shared this belief; instead, she implicitly imputed to Graettinger the “widespread belief’ that others at the plant held. However, the only evidence to which the ALJ points in support of her conclusion is the testimony of Sandra Smith. The ALJ wrote that “Sandra Smith testified that she believed at the time that Corliss Hepburn was the author of the letter, and that she believed this opinion was widespread throughout Plantation.” After a thorough review of the transcript of Sandra Smith’s testimony, however, it is apparent that Smith said nothing of the kind. In reality, Smith testified that (1) she did not know who wrote the letter and (2) “there was a whole lot of gossip going ... ”, at which point her testimony was interrupted. Her earlier reference in her testimony about “gossip” referred to rumors about the letter and union organization in general. Smith said nothing about any “gossip” that Hepburn was supposedly the letter’s author. It is perplexing that the ALJ concluded that Sandra Smith had testified that there was a “widespread belief’ that Hepburn wrote the unsigned letter. That finding was not supported by any evidence, much less substantial evidence.

We now turn to the third and final reason that the ALJ found the element of knowledge regarding Hepburn: “Graet-tinger’s reaction to Hepburn’s mention of the Union after her discharge.” The ALJ reasoned as follows:

Graettinger’s reaction to Hepburn’s mention of the Union after her discharge, his lack of any expression of surprise or any denial, and especially his smile, all support the finding that he was well aware of her union activities. The fact that a manager would smile at an employee whom he had just discharged and who was obviously very upset is inexplicable unless it is interpreted as a smile of triumph responding directly to Hepburn’s remark that she was being fired because of the Union. I find that Graettinger’s smile was, in fact, a response to Hepburn’s accusation that she had been fired because of her union activities. I find, furthermore, that this reaction was an indication that not only was Graettinger well aware of Hepburn’s union activities, but also that he was delighted with the accomplishment of his unlawful action.

In short, the ALJ reasoned that Graet-tinger’s smile, in conjunction with his failure to react or protest in the face of Hepburn’s remark that he fired her because of her union activities, leads inexorably to the inference that Graettinger knew of Hepburn’s protected acts. The problem with this reasoning is that it has no eviden-tiary support. Only Hepburn and Graet-tinger were in on this conversation. Hepburn’s version of the discussion was that the two spoke at the back door of the Plantation facility, as Graettinger was escorting Hepburn from the building. According to Hepburn’s testimony, she said “Ralph, I know why you’re doing this. It’s because of my involvement with the Union.” As Hepburn recounted the situation, Graettinger then “said, no, the Union has nothing to do with it, with a smile on his face.” Graettinger, by contrast, testified that Hepburn had not accused him of firing her for union activities and that he had not smiled at her. There are thus two permissible views of the evidence: (1) Graettinger smiled and denied Hepburn’s accusation that he discharged her for protected activities; or (2) Hepburn did not make the accusation and Graettinger did not smile. Although she supposedly credited Hepburn’s testimony, the ALJ came up with a third scenario, that Hepburn made the accusation and that “Graettinger did not answer, but smiled at Hepburn.” No evidence, let alone substantial evidence, supports that factual conclusion.

Hepburn did indicate that Graettinger smiled, but the ALJ did not reason that the smile, by itself, proved Graettinger’s knowledge; she found that a smile in the context of Graettinger’s silent response to Hepburn’s accusation established knowledge. But it was not a silent response. Hepburn quoted Graettinger as saying, “No, the union has nothing to do with it .... ” Still, the ALJ concluded he lacked any expression of surprise “or denial.” (Emphasis added.) After ignoring his reported denial, the ALJ then concentrated on a motive for the smile. As the attorney for the General Counsel conceded during oral argument, the smile was “not at all critical to her [the ALJ’s] finding.” We nonetheless note that Graettinger’s smile, by itself, would have been too slender a reed upon which to find either knowledge or animus. See Staats and Staats, Inc., 254 NLRB 888, 894-95, 1981 WL 21137 (1981) (holding that the fact that a manager saw an employee smile at a notice for a representation election was not sufficient evidence of that manager’s knowledge of the employee’s pro-union sentiment).

Yet, somehow, the ALJ concluded that “especially his smile” supported the finding that Graettinger was well aware of Hepburn’s union activities. The ALJ characterized it as “a smile of triumph,” indicating that “he was delighted with the accomplishment of his unlawful action.” The ALJ injected all of this psychoanalysis into Hepburn’s simple statement that Graettinger “had a smile on his face.” It is a stretch of the ALJ’s imagination, however, to find that Graettinger’s smile can only be explained by the inference that he knew of Hepburn’s union activities and was “delighted” to have fired her for those acts. People smile for many reasons. In Graettinger’s case, his smile may have reflected consternation that Hepburn had accused him of breaking the law. Or perhaps Graettinger did smile in triumph, because he was happy to be discharging an employee who, in his mind, was a thief. Or maybe he was just nervous, or relieved that the incident was over. See Paradise Post, 297 NLRB 876, 877, 1990 WL 122421 (1990). Since he smiled, according to Hepburn, while saying “No, the union has nothing to do with it,” perhaps he was showing some empathy for the predicament she had created. The fact is, we do not know why Graettinger smiled when Hepburn accused him of violating the law (if, indeed, he smiled at all), and neither does the ALJ. We are certain, however, that the ALJ was incorrect when she reasoned a manager’s smile in such circumstances compels the inference that the manager was aware of union activities. See id. (reasoning that “[t]he fact that Brown was smiling does not compel the inference that Brown was not upset”).

There is an alternate reason that substantial evidence does not support the element of knowledge regarding Hepburn. All three factual predicates upon which the ALJ found knowledge share one determinative flaw: the ALJ did not rely upon substantial evidence indicating that any one of them had occurred before Graettinger decided to fire Hepburn. Knowledge of union activities is relevant only insofar as it allows the factfinder to conclude that the employer’s adverse decision could have been motivated by that knowledge. See NLRB v. Advance Transp. Co., 965 F.2d 186, 191 (7th Cir.1992) (reasoning that the court must determine “if there is substantial evidence to support the Board’s conclusion that the General Counsel met his burden of showing by a preponderance of the evidence that Advance’s decision to terminate Tuffs and Bauldry was motivated in any way by animus toward their protected activity”). Thus, where substantial evidence shows that the decisionmaker learned of the employee’s union activities on Monday, and decided to fire that employee on Tuesday, the ALJ could find that the element of knowledge existed; conversely, where the decisionmaker decided to fire the employee on Monday, but substantial evidence shows that he did not learn of that worker’s union activities until Tuesday, the element of knowledge could not be met, because a decisionmaker cannot be motivated by what he has yet to learn. This case is analogous to the latter scenario.

The ALJ found as a matter of fact that Graettinger “immediately sought permission to discharge” Hepburn and Richardson as soon as he learned about Richardson’s incorrect report that a service flash covered Hepburn’s refrigerator. The ALJ did not make a finding as to the particular day on which Graettinger “immediately” decided to fire the two. The evidence, which the ALJ did not address, shows that Graettinger almost certainly learned of the service-flash incident by February 19, 1998. Richardson’s supervisor, Pat McLaughlin, testified that he had discovered Richardson’s false report on “either the 18th or 19th, more than likely” and that he and support manager Horacio Villazon then reported the problem to Graettinger on the same day that he uncovered it. Villazon, for his part, testified that McLaughlin had alerted him about Richardson’s false report on February 19 and that they had reported the matter to Graettinger on the same day. There is, as far as we discern, no contrary evidence in the record. We now turn to the dates on which, according to the ALJ, Graettinger learned of Hepburn’s protected activity.

First, there is the company-mandated antiunion meeting at which Hepburn spoke in favor of the Union. Without discussing any of the evidence as to the meeting’s date, the ALJ found that meeting occurred in “mid-February.” “Mid-February” is an elastic term that does not really tell us whether the meeting occurred before February 19. Most of the witnesses who testified as to the meeting’s date, however, recalled it as having been within one week of the representation election, which was held on Friday, February 27. Hepburn testified that the meeting “was the week before the election.” Graettinger testified that the meeting that Hepburn attended was held “about mid week” during the week of February 27. Joe Hofer, a co-worker of Hepburn’s and a fellow supporter of the Union, testified that the meeting was held within a week before the election. The only evidence to the contrary, as far as our own review of the record indicates, is that of Cordy Richardson. Richardson testified that the meeting was in early to mid-February.

Had the ALJ explained why she found credible the part of Richardson’s testimony that placed the meeting in mid-February and, arguably, before February 19 (as opposed to the part of his testimony suggesting that the meeting was in early February), and why she found incredible the testimony of Hepburn, Hofer, and Graettinger that the meeting was held no more than a week before the election (and thus after February 19), we would likely hold that substantial evidence supported the conclusion that the antiunion meeting happened in mid-February, and prior to Graettinger’s decision to fire Hepburn. As we stated earlier, the ALJ’s credibility determinations are entitled to particular deference. The ALJ, however, never explained why she rejected the weight of the evidence and found that the antiunion-meeting happened early enough to have influenced Graettinger’s decision to fire Hepburn. That is not good enough. Even when the record contains some evidence that could conceivably have supported an ALJ’s finding, the substantial evidence standard is not met if the ALJ does not discuss, or even provide a citation to, that evidence. Scivally v. Sullivan, 966 F.2d 1070, 1076 (7th Cir.1992) (holding that “the ALJ must minimally articulate his reasons for crediting or rejecting” evidence). We hold that no substantial evidence supports the ALJ’s implicit finding that the February meeting occurred early enough to have influenced Graettinger’s decision to discharge Hepburn.

We next consider the unsigned letter that Joe Fowler circulated. Although the letter is dated February 20, the ALJ found that it was “widely circulated at Plantation in mid-February.” The only evidence the ALJ mentions regarding this conclusion is Fowler’s testimony that he had drafted the letter “a few weeks before the election.” That Fowler drafted the letter a few weeks before the election (which was held on February 27), however, does not lead to the conclusion that the letter was necessarily distributed before Graettinger decided to fire Hepburn (by February 19). The evidence, as far as we can discern, supports the contrary conclusion, that Fowler’s letter was not distributed until the week before the election (which, as we have discussed, was almost certainly after Graettinger had made up his mind to terminate Hepburn’s employment). The letter itself is dated February 20, 1998. Hepburn testified that she received the letter “at or around the time” of the mandatory antiunion meeting. There is, as far as we can tell, no substantial evidence showing that Fowler’s letter was circulated early enough in February to have influenced Graettinger’s decision to fire Hepburn.

We next address the third and final reason that the ALJ found knowledge regarding Hepburn: Graettinger’s reaction to Hepburn’s accusation of anti-union animus immediately after he officially fired her on March 6, 1998. If Graettinger decided to fire Hepburn “immediately” upon learning in February of the service-flash incident, then nothing Hepburn told him about her protected activity in March could have influenced his decision. To the extent that Graettinger’s reaction could reasonably be construed as an admission that he had known of Hepburn’s protected activities all along, it would constitute substantial evidence of knowledge. Here, however, there is no evidence that Graet-tinger could have known of Hepburn’s union activities until March 6, which was almost three weeks after he had made the decision to terminate her employment, and the reaction of Graettinger to Hepburn’s accusation is too amorphous to support the inference that he knew of Hepburn’s union activity when he made the decision to terminate in February. As discussed above, Graettinger’s smile and denial that he had fired Hepburn because of the Union does not constitute substantial evidence that he knew of Hepburn’s protected acts when he made his decision on February 18 or 19.

The ALJ did not make findings of fact, grounded in the record, as to the chronological relationship between Graettinger’s decision to fire Hepburn and the events that (in the ALJ’s mind) showed that Graettinger knew of Hepburn’s union activities. Had she done so, she probably would have realized that those three events happened (if they happened at all) only after Graettinger decided, on February 18 or 19, to discharge Hepburn “immediately” upon learning of the service-flash incident. For the purposes of this appeal, it is enough to say that the ALJ did not rely on substantial evidence showing that Graettinger knew of Hepburn’s union activities when he decided to fire her. Our own review of the evidence indicates that Graettinger probably was ignorant of Hepburn’s protected acts, or at least the three ostensibly-protected acts that the ALJ identified, when he discharged Hepburn. Because Graettinger’s decision to fire Hepburn could not have been motivated by what he had yet to discover, none of the three reasons that the ALJ gave for concluding that Graettinger knew of Hepburn’s union activities is substantial evidence of the element of knowledge.

C. Cordy Richardson

We turn finally to Cordy Richardson, once again focusing on the element of knowledge. Because the Board adopted-the ALJ’s findings and analysis as to this issue, we review the ALJ’s opinion. The ALJ found the element of knowledge for three reasons.

[H]is attendance at the same meeting and his overt agreement with Hepburn’s remarks there, are direct evidence that he was believed by Respondent to be a Union supporter. Charlie Young’s knowledge that Richardson supported the Union is imputed to Respondent. In addition, his well-known association with Joe Fowler is circumstantial evidence that Respondent believed him to be a Union supporter. On this subject, the General Counsel has urged that the March 1997 surveillance of Fowler, Richardson, and others at the lunch hour was in fact surveillance of their union activities. While I find that evidence of the purpose of McLaughlin’s binocular viewing of the lunch group is insufficient to find that he was surveying their Union activities, I find it does establish he knew Richardson and Fowler, the main Union activist, were associated together. This is circumstantial evidence which supports the inference that Respondent believed Richardson supported the Union, especially when viewed in conjunction with the other evidence tending to the same conclusion.

As we discussed above in relation to Hepburn, Richardson’s attendance and comments at the mandatory meeting are not sufficient evidence of knowledge because the ALJ relied on no substantial evidence that the meeting and remarks occurred before Graettinger had made up his mind to discharge Richardson and Hepburn. In fact, as we discussed above, it is likely that the meeting occurred after Graettinger had determined to fire Richardson. Richardson’s attendance and remarks at the meeting, accordingly, are not substantial evidence of knowledge.

We turn next to the second reason that the ALJ found knowledge as to Richardson, her conclusion that “Charlie Young’s knowledge that Richardson supported the Union is imputed to Respondent.” We emphasize again that Sears’s knowledge of Hepburn’s protected conduct or speech, for purposes of the prima facie case, can only be imputed through the decisionmaker, Graettinger. The question then is whether Young’s knowledge of Richardson’s protected comments could be imputed to Graettinger. The ALJ mentioned no evidence that would lead to the conclusion that Young told Graettinger about Richardson’s “overt agreement with Hepburn’s remarks” at the meeting; all we are left with is her conelusory decision to impute Young’s knowledge to “Respondent.” Accordingly, we hold that substantial evidence does not support the imputation of Young’s knowledge to Graettinger.

We turn to the final reason that the ALJ found knowledge: Richardson’s “well-known association with Joe Fowler.” This reason is deficient on two counts. First, the relevant knowledge, as we discussed above, is knowledge that the employee engaged in protected activities. The ALJ discussed no evidence showing that, when Richardson ate lunch with Fowler, the two were involved in conduct within the ambit of § 157. As far as we can tell from our review of the ALJ’s opinion, they were just eating lunch. Even if Graettinger had been aware of that rather unremarkable behavior, therefore, it could not be said that he knew that Richardson was involved in protected activities.

The second reason that knowledge may not be based on this evidence is that, even if the mere act of dining with a union activist could somehow be considered protected conduct, the ALJ discusses no evidence that Graettinger was aware of that behavior. The ALJ mentions “McLaughlin’s binocular viewing” of the restaurant where Richardson and Fowler dined. However, she discusses no evidence showing that: McLaughlin saw Richardson with Fowler; Richardson was even present on the day of McLaughlin’s surveillance; or, that McLaughlin had ever told Graettinger of Richardson’s association with Fowler. Although there is substantial evidence that Richardson often ate lunch with Fowler, there is no substantial evidence that Graet-tinger was aware of that fact when he decided to end Richardson’s employment.

The ALJ identified no substantial evidence showing that Graettinger knew of Richardson’s protected activities before he decided to fire Richardson. We therefore deny enforcement of the Board’s order as to Richardson as well.

III.

The primary lesson of this case is that facts matter. In many instances, the Board or ALJ made important findings of fact that were unsupported by substantial evidence or were clearly incompatible with the record. In still other instances, the Board or ALJ simply ignored strains of evidence that did not mesh with their ultimate conclusions. On yet other occasions, the Board or ALJ failed to make crucial findings of fact. It is difficult, in light of these discrepancies, to avoid the conclusion that the Board and ALJ were striving to reach a predetermined result. For our part, having viewed the record in its entirety — including the body of evidence opposed to the Board’s view — we cannot conscientiously find that the evidence supporting the Board’s decision is substantial. We therefore grant Sears’s petition for review and deny enforcement of the Board’s order.

1 . The Board, without discussing the evidence on which it based its conclusion, found as a matter of fact that Iaci distributed authorization cards and discussed the Union with coworkers in the parking lot "between May and June” 1997. On appeal, the General Counsel’s position is that these acts actually began in April 1997. The only evidence to which any party cites as to when Iaci engaged in this conduct is Iaci's own testimony to the effect that the time period was some time in April or May 1997. We are therefore satisfied that substantial evidence supports the conclusion that Iaci’s union activities began as early as April 1997 and ended by June 1997.

2 . The testimony of Iaci and Gonzalez differed as to whether Gonzalez told Iaci to add anything to his statement. The Administrative Law Judge (''ALJ”) found that "Gonzalez insisted that [Iaci] add the last two sentences to his 'statement,' ” and the Board adopted that factual finding. Because substantial evidence supports that finding, we assume it to be correct for the purposes of our review that Iaci did not actually admit to the last two sentences in his statement.

3 . The ALJ, without discussing or citing any evidence, placed this meeting in "mid-February.” As we discuss in detail later, the weight of the evidence shows that the meeting took place at some point during the week of Monday, February 23, 1998.

4 . The Board found as a matter of fact that the election was held on Saturday, February 28, 1998. In his brief, the General Counsel states that the election was held on Sunday, March 1, 1998. However, the uncontested evidence, in the form of the NLRB’s tally of ballots, shows that the election actually was held on Friday, February 27, 1998. As will become evident, the date of the election is important.

5 . Gonzalez denied that he told any employee what to write. But the employees testified that he had done so, and the ALJ found their version of events more credible because "Gonzalez’ memory, manner of testifying, and demeanor were entirely unconvincing.”

6 . During oral argument, the attorney for the General Counsel conceded that “Graettinger is the decisionmaker ... Graettinger is the one who fired Iaci.... The question needs to be whether or not he had bias.”

7 . Although we need not resolve the matter in this case, we note that it is less clear whether the timing of an employee’s discharge, by itself, may constitute substantial evidence of antiunion animus. Some authorities hold that it may not. See Chicago Tribune Co., 962 F.2d at 717-18 (reasoning that "mere coincidence is not sufficient evidence of antiunion animus”); NLRB v. Loy Food Stores, Inc., 697 F.2d 798, 800-01 (7th Cir.1983) (reasoning that evidence that the employer fired a worker while he was involved with an organizing campaign was not substantial evidence of the prohibited animus). At least one opinion holds that "[t]iming alone may suggest anti-union animus.” NLRB v. Rain-Ware, Inc., 732 F.2d 1349, 1354 (7th Cir.1984). Still another authority takes a middle view, reasoning that “mere coincidence alone, without other circumstantial evidence, may not always support an inference of animus.” Martech Med. Products, Inc., 331 NLRB 487, 501, 2000 WL 888355 (2000).

8 . In general, a decision to discharge an employee cannot be caused by an antiunion animus that the decisionmaker does not possess. We nonetheless note that there may be limited circumstances in which it is proper to impute the animus of a non-decisionmaker to the employer. In Grand Rapids Die Casting Corp. v. NLRB, 831 F.2d 112 (6th Cir.1987), the Sixth Circuit held that the antiunion animus of a supervisor could be attributed to the company because, even though that supervisor was not the decisionmaker, he knew of the employee’s protected activities and was involved in the decision to fire the employee. Vulcan, 219F.3dat686 (citing Grand Rapids, 831 F.2d at 117). In this case, however, Medford was not involved in the decision to discharge Iaci, and his animus is totally irrelevant.

9 . Inexplicably, the ALJ wrote that "Christine Smith testified that she had no problems with [Iaci] during the several months prior to August 1997.” The transcript of Smith’s testimony, as we have discussed at length, shows otherwise. Smith’s testimony was that, during the several months before August 1997, she had documented numerous instances in which Iaci had filed false reports. Smith also testified that she had told Iaci ”[t]hat he needed to make sure that doing the ethical thing” was the way that he improved his performance record.