Johns Hopkins Magazine - April 1995 Issue

Science & Technology

Top quark, etc.

At long last, the elusive top quark is found

After 20 years of searching for the subatomic particle known as the top quark, Hopkins particle physicist Bruce Barnett can reap his reward. In early March, Barnett and a large international team of collaborators announced that they had discovered this last fundamental unit of matter.

"This is very exciting," said a beaming Barnett. The professor of physics and astronomy led the Hopkins team that helped design one of the key instruments used to detect the top quark. It is called the silicon vertex detector.

The discovery confirms the Standard Model of physics, says Barnett, which states that all matter is made up of six different subatomic particles called quarks--named up, down, charm, strange, top, and bottom--and six different particles called leptons. Until now, physicists had proven the existence of 11 of the particles. The top quark, which exists for just a trillionth of a trillionth of a second before disintegrating into other particles, was the last holdout. "I expect this is the most difficult physics analysis ever done," says Barnett.

The discovery was made at the Fermi National Accelerator Laboratory, outside of Chicago, where Barnett and his teammates have been conducting experiments since 1989. The physicists create top quarks by smashing together protons and their oppositely charged counterparts, antiprotons, at super-high speeds. The collision yields jets of particles (including, very occasionally, a top quark) that are identified by three detectors including the silicon vertex detector.

The detector contains four layers of silicon covered by thin aluminum strips. When particles pass through the silicon they knock out electrons, which jump to the aluminum strips. Using electronics attached to the device, the physicists identify where the particles hit each layer of aluminum, which tells them the path the particles took coming into the detector. They trace the path backward to initial collision to determine whether it has the trademarks of a quark-producing event.

On Halloween 1992, the physicists detected their first "event" indicating a top quark had been produced. They continued to collect data, and exactly a year ago, announced they had observed 12 events, enough to claim they had "evidence" for the top quark. They then fine-tuned their experiment and continued to collect data. By now they have observed more than 50 events-- enough to conclude, says Barnett, "This is a discovery."

The discovery is not an end but a beginning, he says. Now physicists can explore whether the top quark behaves the way theory says it does. One puzzle is the mass of the particle. The recent experiments indicate that the top quark weighs 176 billion electron volts, more than any other particle, and 10 times what scientists had predicted a decade ago. --MH

The how and why of word stress

How do we generate new words out of old ones, without generating confusion? Polite to impolite, form to formative, divide to dividing--those are easy. But in other cases, you'd think we'd be confused. Consider:

It all seems very arbitrary. As one word transmutes into another, vowels turn long or short. Syllables mysteriously swap emphasis. In many cases, as in IdEa, ide-Ation, the old root could be a different word altogether, so different is the word stress. Yet, points out Hopkins linguist Luigi Burzio, in our native languages, we all can follow such transmutations with ease. What he wants to know is, how do we do it?

When in doubt, make a rule: that was the old approach. It seemed that we must all have learned a set of implicit rules, from experience, about the various prefixes and suffixes. So for decades linguists beavered along to find them, piling up masses of special words, groups of words, and situations, each one requiring its own special rule, with of course exceptions.

The structure was--well, call it unruly. Too complex to be true, given how easily children learn their native languages, no special gift required. To modern linguists, it followed that language learning must build on some initial structure, something you might call the language faculty.

That's why, explains Burzio, linguists now puzzle over language primarily for clues about the "cognitive structures, the sets of word-knowledge that the human brain innately has." (It's no accident that Burzio is a professor in the Department of Cognitive Science.) Furthermore, they expect these structures to have a certain elegant simplicity, like that of DNA: a few simple conditions that generate endless variety.

Which is why Luigi Burzio is a happy man. He has discovered two conditions, just two, whose interaction he believes explains it all: how word stress shifts when nouns and verbs take affixes (either pre- or suf-) to form new words. "It's extremely regular," he exults.

Briefly, one principle is that stem stresses must be preserved, as in

In the second case, Burzio argues that mag retains stress, but that the stress is now secondary; primary emphasis shifts to ca. (Similarly, NapOleon, NapOleonic.)

It is easy to see how preserving stem stress would help keep speech understandable. And the principle also, says Burzio, clarifies such puzzles as person, personification: the new word does preserve stress--but not of the root. Rather, it remembers the intermediary word, personify. (More about personify later.)

Burzio's second key principle is that when an affix of any sort is added, a long stem vowel must shorten, as in

Here both new words bid farewell to long E and I, while also preserving stem stress. In many cases, says Burzio, the two principles work together.

And when they don't--it's one or the other. DesIre, desIrous, for example, preserves the stress but does not shorten I, while in exclAim, exclamAtion, the root's A shortens but loses its stress.

The underlying requirement, he says, is metrical: Whatever happens, words must have well-formed metrical "feet": Syllables must fall in particular groupings that are specific to the language.

That's why, in English, personify doesn't preserve the stress of person. It cannot. Try to say it as if it did-- PER son, PER so ni fy. Hard, isn't it? The English language simply does not allow three tiny syllables like so ni fy to scrabble along with no emphasis. To have stress, there must be contrast, explains Burzio.

He argues, in fact--and this is one of his innovations--that English words have only two basic foot structures. One is bisyllabic, as in

The second foot structure has three syllables:

Do not be confused by the initial a, ba, and per in those examples: they're just leftovers, explains Burzio. In scanning poetry, you'd have to account for them, but from any linguist's point of view--he shrugs--"You just ignore it."

That said, you can see that the stress of any foot falls on the leftmost syllable, while feet are built from the end of the word.

In Burzio's scheme, the many words like e(mOte) and ex(press) are bi-syllabic: the unaccented e and ex are throwaways, and the strong final consonants have "something like a null vowel. In those cases," explains Burzio, "a final empty structure is not uttered, but mentally represented." (Say the words out loud: emote, express. You can almost hear it, a kind of silent -uh as you finish the word.)

Similarly, words we normally consider to have one syllable-- form, hit, cat--also have final empty structures, he says. That makes them, too, bi-syllabic.

In fact, he has concluded that in English, no foot can consist of a single sound. "If a unary foot were possible," he argues, "at least sometimes you ought to be able to say something like ex PLA NA tion [with ex a throwaway, PLA the unary foot]. That would preserve stress, so you'd think it would be possible. But," he says with emphasis, "we never find this!"

Burzio sits back and smiles. "Our brains were born knowing that there were no unary feet." He finds none in Italian, Latin, Hebrew, Polish, or Spanish, and he doubts they exist in any language. How could they? Stress is key to meaning, and it only exists through contrast. There can be no front without a back, no stress without something unstressed.

Cross-language study is not his field, however, and Burzio is more than content to have crunched dozens of rules into one unified system, for English: When a noun or verb takes an affix, the original sound is preserved as much as possible, within the requirements of well-formed feet, by stress preservation, vowel shortening, or both. A book explaining the work, Principles of English Stress, was recently published by the Cambridge University Press and has received much attention among linguists.

"Language," says Burzio, "is a generative device that can create an endless series of ideas with relatively few sounds." It sounds like DNA, doesn't it? --EH

Virtual bronchoscopy, and more

Jose Ma. Maisog, an investigator with the National Institute of Mental Health, wears 3-D glasses to view the new Electronic Brain Atlas System, a digital collection of 3-D MR brain images that will be a reference tool for radiologists and neurosurgeons. By comparing a patient's MRI to the atlas, doctors will pinpoint malfunctioning regions.

This technology, with others, was demonstrated in February at the School of Medicine, where researchers from the Center for Information-enhanced Medicine (CIeMed) showed off technology they say will take medical imaging into the 21st century. The center, a joint venture between the School of Medicine and the National University of Singapore's Institute of Systems Science, is developing graphics tools and programs that expand the usefulness of techniques such as magnetic resonance imaging (MRI) and computed tomography (CT).

Another technique that was unveiled, virtual bronchoscopy, allows radiologists to "walk" down the airways of a breathing lung in search of mucus or blood clots. Unlike conventional bronchoscopy, in which a tube is snaked down the patient's throat, this technique is non-invasive. The group also showed a video movie of surgeons "probing" a beating human heart for signs of disease. They were actually manipulating 3-D MR images displayed on a computer screen. The non-invasive procedure, said conference organizers, would cost about a fifth of what the current set of tests does.

All the techniques are still in experimental stages, says Carey Kriz, director of CIeMed. But considering that the Hopkins/Singapore partnership is just a year old, "applications are moving ahead rapidly." --MH

Written by Elise Hancock and Melissa Hendricks

Send EMail to Johns Hopkins Magazine

Return to table of contents.