Publishable definitions of things are hard to come by, but an older definition of engineering is that it was the art of managing engines. Which certainly points out how the people who drive trains got to be called engineers, and also the guild of "operating engineers". Neither of these groups "practice engineering" as it is typically laid out in legislation around the world.
The modern (circa 2003) engineer typically has an extensive background in mathematics and science. An engineer is typically called upon when routine skills of practise are insufficient to solve the technical problem at hand. With knowledge increasing at such a rapid (and increasing) rate, this is a common occurance. Routine skills are learned by repetition, work in new areas has had no chance for anyone to utilize repetition.
At the end of the day, the engineer is charged with applying this advanced knowledge to the benefit of society in a safe way, for the benefit of human welfare. What level of technical knowledge is required of the engineer, and where the dividing line is between routine skills of practice and engineering is fluid. These qualities change with time, and location (through the average and individual characteristics of the people at that location).
Some people think that engineers are elitist. Certainly some members of engineering constitute an elite group, however I doubt that few would think that engineers should rule society. Typically it is society as a whole that determines the scope of engineering, and not the engineers themselves.
In terms of technical knowledge, scientists typically have the required amount of knowledge, especially technical knowledge, to participate in its application (in terms of some kind of engineering work). Typically, the scientist is concerned with the pursuit of new knowledge, and not the useful (and safe) application of existing knowledge. But scientists and engineers are both professionals, and both are expected to only pursue work they are qualified to perform.
Where the highest chance of conflict is, in the boundary between the trades and engineering. Typically, the "proof" by a tradesperson, that a tradesperson can or cannot accomplish a task safely is in the performance of the task, and can only be determined after the fact. It is possible that an engineer (or possibly a scientist) can prove before the fact that a task is possible to complete safely or not. A problem with this situation is that the tradesperson often doesn't have the technical background to understand the proof. One possible result is that the tradesperson thinks an engineer's ruling constitutes a conflict of interest, and not trying to prevent an unsafe condition from getting into the public realm.
Another concept contributing to this conflict, is that the dividing line is a function of the skills of the tradesperson. So, if the engineering profession "allows" one tradesperson to perform a certain task, they may disallow another tradesperson from performing the same task.
The conflicts which can look like restraint of trade from one point of view, often rely in part on the professional ethics of the engineer(s) involved. The engineer(s) in question shouldn't interfere just to restrain trade. In extreme cases, the only resolution for the matter is to ask another profession (the law profession, or in extreme cases ask the government to introduce new legislation or edit existing legislation) to rule on things.
The public deserves the right to spend as little as they have to, to purchase things they need or want. Many common services of a technical nature are simply easier and cheaper to procure from tradespeople. As long as they (tradespeople) can do so in a safe manner, they should be allowed to do so. With experience, tradespeople should recognize where the services of an engineer are required, and they should recommend that an engineer be consulted in those circumstances. Likewise, if a task doesn't require the services of an engineer, the engineer should recommend a cheaper alternative which may include the services of a tradesperson. This is just professional courtesy.
If an engineer is providing services in a regime close to where tradespeople commonly work; they should endeavor to produce results that keep to the spirit of the common practices the tradesperson uses in that close regime. If such common practices exist. Common examples of this, are things like the building code or the electrical code. Future repairs, renovations or adjustments on part of the engineers work may become desirable, and if the task was done in the spirit of the nearby common practice, it may be possible to effect these changes without invoking the services of an engineer. If the engineer feels that future changes would need further engineering input; there should be some way to indicate this, including non-standard practices.