United States vs. Martha Stewart
Mightier Than Broker’s Word, 2 Ballpoints Could Help Land Martha Stewart in the Pen
Updated July 1, 2003 12:01 a.m. ET
Will Martha Stewart be undone by a blue ballpoint pen?In the case against Ms. Stewart, a key piece of evidence is a tiny, handwritten notation made by her stockbroker on a trading worksheet filled with similar scribbles. Prosecutors claim the broker belatedly inserted the note to help cover up Ms. Stewart’s improper stock trading. Their support: Laboratory analysis showing that the blue ballpoint ink he used is different from ink elsewhere on the document.Forensic ink analysis, a little-known crimefighting tool, is suddenly in the spotlight in several high-profile business scandals. In Harrisburg, Pa., federal prosecutors pursuing accounting fraud at Rite Aid Corp. used ink analysis to help show that certain documents were backdated. And in San Francisco, the CEO of Aptix Corp. has been indicted on federal charges that include perjury, after ink analysis and other evidence indicated he had fabricated a notebook to help win a patent lawsuit. He has pleaded not guilty.
Scrutinizing ink is on the rise, in large part, because investigators are spending more time chasing corporate officials for crimes that can be difficult to prove or that are mind-numbingly complex. In many cases, it can be easier to demonstrate a coverup than the underlying crime. For example, the government charged Ms. Stewart and her broker, Peter Bacanovic, with perjury and obstruction of justice, not insider trading. And what better way to prove a coverup than with scientific evidence of fabricated or altered documents?
“The issue in a lot of these white-collar cases isn’t the substantive offense,” says Nick Theodorou, a former federal prosecutor now at Foley Hoag LLP in Boston. “It’s when they panic, especially control freaks, and they alter things. That’s when you get into obstruction-of-justice problems.”
Ms. Stewart and Mr. Bacanovic have pleaded not guilty. Ms. Stewart’s attorney declined to comment. Mr. Bacanovic’s attorney, Richard Strassberg, said, “Mr. Bacanovic did nothing wrong and that will be proven at trial.”
Most of the ink analysis for the federal government — including analysis in the Rite Aid case — is done by the U.S. Secret Service’s Forensic Services Division, which maintains the world’s largest library of ink samples with more than 7,000 entries. In addition, there are a handful of private experts around the country who do ink analysis, mostly for medical-malpractice suits, patent battles, or cases involving disputed wills.
To identify a particular type of ink, scientists typically use a technique called thin-layer chromatography. A tiny sample of ink is removed from the paper using a hypodermic needle, then dissolved in a solution and applied to a thin plate. The plate is put through a chemical process in which the component dyes start to separate.
Albert H. Lyter III, an ink expert in Raleigh, N.C., says a complicated ink can be made of as many as seven or eight different dyes, although a typical one has just three or four, the most common being Methyl Violet and Victoria Blue. There are at least 100 different dyes, from the red-tinted rhodamine family to gray-hued nigrosine.
After separating the dyes, scientists try to match the ink with known formulations. Mr. Lyter says Bic Corp.’s black pen is an easy one, because it contains only two main dyes and is so commonly seen. Others can be trickier, he says, because certain ink formulas are very similar, and even the best library doesn’t contain every ink sample.
One early case in which ink science played a role was the 1973 conviction of mass-murderer Juan Corona, who hacked to death 25 migrant workers near Yuba City, Calif. Although Mr. Corona denied involvement in the killings, a key piece of evidence was a ledger book with the names of seven victims found in Mr. Corona’s home. Richard Brunelle, an ink-analysis pioneer then working for a predecessor to the Secret Service lab, recalls that he was able to match the ink in the ledger to a pen found in Mr. Corona’s possession.
Ink analysis even figured into a 1993 episode of the television sitcom “Cheers.” Minutes before an Internal Revenue Service audit, Norm starts scribbling fake diaries and receipts. After the IRS agent, played by an attractive woman, threatens to send the paperwork for analysis to see whether the ink was more than 30 minutes old, a panicked Norm tries another technique: seducing the agent.
In the Rite Aid case, former Chief Executive Martin Grass and his lieutenants allegedly used a slightly more elaborate plan, creating a raft of backdated letters as part of the plot to cover up their accounting misdeeds. In pretrial documents, prosecutors said the Secret Service lab had identified the ink used by Mr. Grass to sign one of the letters. That particular ink, they said, wasn’t commercially available until three months after the letter was dated.
Last month, just before he was to face trial, Mr. Grass pleaded guilty to two conspiracy counts, including obstruction of justice, and admitted his role in preparing the backdated letters.
In the Martha Stewart case, the analysis task was simpler — differentiating between two inks on a single document. When investigators started asking questions about Ms. Stewart’s sale of ImClone Systems Inc. shares just ahead of negative news, Ms. Stewart claimed that she had a pre-existing arrangement with her broker, Mr. Bacanovic, to sell the shares if they fell below $60. To support that claim, prosecutors say, Mr. Bacanovic added a tiny “@60” beside the ImClone name on a worksheet on which he had previously jotted notes about planned sales of Ms. Stewart’s other holdings. The notation, prosecutors claim, was added sometime in the five weeks after the trade.
Erich Speckin, of Speckin Forensic Laboratories in Okemos, Mich., says determining that two separate inks were used on the worksheet would be “one of the easiest things” for an expert, much more certain than identifying the particular inks used. Mr. Speckin doesn’t know what test was used — and the government isn’t saying — but he speculates that the government lab probably didn’t use thin-layer chromatography, because the “@60” mark is so tiny that taking multiple samples would destroy the evidence. Instead, he suspects the government used a type of infrared imaging. That test, he says, “would be definitive” if it showed a difference between the “@60” mark and others on the page.
A person close to the case says Ms. Stewart’s lawyers are likely to press the government to prove that the notation was added later, not just that a different pen was used. This person says it’s possible Mr. Bacanovic could have been interrupted when originally marking up the document and had simply picked up a different pen. “What matters is the time period, not the pen,” the person adds.
Mr. Speckin says ink analysis probably won’t be able to help with that. Because ink soaks into paper slowly and various components evaporate at different rates, scientists can use extraction tests to date many ink samples. Mr. Speckin likens it to trying to remove a stain on clothing: The longer it has been on, the less of it will come out. But Mr. Speckin says the tests are only fine enough to differentiate between samples six months apart, not the five weeks or less alleged with Mr. Bacanovic’s worksheet. (Ballpoint ink dries completely after several years, so the process doesn’t work on older documents.)
Documents generated with inkjet printers can be scrutinized using the same techniques. But matching an ink sample to a particular printer is impossible, and even matching it to a brand of printer can be impossible in some cases, because there are so many companies around the world making printer inks or refilling cartridges. Documents printed on laserjet printers are even trickier, in part because the “ink” used is really a type of plastic with a colorant, which is tough to remove from paper and has fewer components to analyze.
What advice would experts give a fraudster intent on covering up document alteration? “Use pencil,” says Mr. Lyter, who calls it “the most difficult thing we have to deal with.” In part, that’s because pencils don’t leave much on the page to analyze, and the basic No. 2 pencil hasn’t changed much over time. Mr. Lyter says it’s possible to tell if two different pencils were used, but identifying the type of pencil, or when a particular mark was made, is difficult.
Reference Link: http://www.wsj.com/articles/SB105701075148372400