Scientists once believed that nothing could travel faster than sound -- hence the notion of a "sound barrier." Investors, these past 14 years, have come to think that stocks can't go much higher than 1,000, as measured by the Dow Jones Industrial Average. That's the "psychologial barrier."

But if we've learned anything about the market during these long hard years, it's that as soon as we all know something for sure, it's going to be wrong. Veteran forecaster Edson Gould thinks we're going to be wrong pretty soon.

Gould is the author of Findings & Forecasts, a high-priced monthly investment report published by Anametrics here. In a special issue called "The Sign of the Bull," he predicts the start of a long-term upward trend in stock prices, of the sort not seen since 1966.

His forecast, in a nutshell: an upward move, over the next few years, to 1,200 to 1,400 on the Dow; an eventual move to 2,500 to 3,000 by the mid-to late-1980s -- a 275-percent gain from 800 in the next 10 years.

Gould is a technical analyst, which means that he forecasts largely on the basis of historical stock-price trends, and little on the politics of the day.

He has a solid record of good forecasts behind him. In 1963, with the Dow at 665, he wrote a special report called "The Next 400 Points," predicting a rise to 1,066 within three years. He missed the top by just 15 points.

In October 1972, with the market at 940, he published "The Last 100 Points." It forecast a short, final rise in stocks, then a 400-point collapse by 1974. In fact, the Dow stocks topped out at 1,051 in January, then went into a free fall, shedding 474 points. In late 1974, he predicted the upturn.

Not to say that he's always right. He got optimistic again in December 1978, which in restrospect was too early a call for the market as a whole.

Regardless of how high the market eventually rises, there are good reasons for thinking an upturn is near. When recessions take hold, interest rates start to fall and stocks traditionally undergo a broad, upward move. After the first few months, selectivity sets in, as investors search out the companies with better earnings.

To Gould's technical eye, the backing and filling of stocks in recent years resembles other inconclusive market periods, which suddenly exploded in major new bull markets.

He sets great store by a measure he calls the Sentimeter, which charts the ratio of stock prices to their dividends. In studies reaching back to 1871, a Dow price of $15 for every $1 of dividends indicates a practically riskless market. Only twice in history had the ratio dropped to $10 -- at the market bottoms of 1917 and 1932. The high-risk level is $30. Today the Sentimeter registers about $16.

Gould's conclusion: "After more than 13 years of a do-nothing market, the years ahead are quite likely to produce a new and exciting period" for stocks.